Article Text

Extended report
Treating to target matrix metalloproteinase 3 normalisation together with disease activity score below 2.6 yields better effects than each alone in rheumatoid arthritis patients: T-4 Study
  1. Yukitomo Urata1,
  2. Ryoko Uesato2,
  3. Dai Tanaka2,
  4. Yoshihide Nakamura3,
  5. Shigeru Motomura4
  1. 1Department of Rheumatology, Seihoku Chuo Hospital, Gosyogawara, Japan
  2. 2Department of Orthopaedic Surgery, Seihoku Chuo Hospital, Gosyogawara, Japan
  3. 3Departments of Orthopaedic Surgery, Hirosaki University Graduate School of Medicine, Hirosaki, Japan
  4. 4Department of Pharmacology, Hirosaki University Graduate School of Medicine, Hirosaki, Japan
  1. Correspondence to Yukitomo Urata, Department of Rheumatology, Seihoku Chuo Hospital, 41 Nunoyacho, Gosyogawara 037-0053, Japan; yurata{at}dream.com

Abstract

Objectives To assess whether therapy to achieve both a disease activity score in 28 joints (DAS28) less than 2.6 and matrix metalloproteinase (MMP) 3 normalisation offers better outcomes than either target alone in early rheumatoid arthritis (RA) at 56 weeks: Treating to Twin Targets (T-4) Study.

Methods 243 early RA patients were randomly allocated to one of four strategy groups: routine care (R group; n=62); DAS28-driven therapy (D group; n=60); MMP-3-driven therapy (M group; n=60); or both DAS28 and MMP-3-driven therapy group (twin; T group; n=61). Medication was started with sulfasalazine (1 g/day) in all intervention groups. Targets were DAS28 less than 2.6 for the D group, MMP-3 normalisation for the M group and both DAS28 less than 2.6 and MMP-3 normalisation for the T group. If the value in question did not fall below the previously measured level, medication was intensified, including methotrexate, other disease-modifying antirheumatic drugs and biological agents. Primary, secondary and outcome measures consisted of the proportions of patients showing clinical remission (DAS28 <2.6), radiographic non-progression (Δmodified total Sharp score ≤0.5), normal physical function (modified health assessment questionnaire score 0), or comprehensive disease remission defined as the combination of clinical remission, radiographic non-progression and normal physical function.

Results Clinical remission at 56 weeks was achieved by more patients in the T group (56%) than in the R group (p<0.0005) or M group (p<0.0005).

Conclusions Results of the T-4 Study reveal that a twin target strategy can achieve a high clinical remission rate in early RA.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Rheumatology has experienced marked advances during the past decade. This development, attributable largely to the introduction of biological agents,1,,4 is not restricted to the growing variety of clinically efficacious targeted therapies,2 but has also resulted in a completely new approach to the management (tight control) of patients with rheumatoid arthritis (RA).5,,14 The concept of tight control—intensive disease management based on specific clinical variables—is well accepted in the management of many chronic conditions such as diabetes,15 hypertension16 and coronary artery disease,17 and has improved patient outcomes. Targets for the treatment of RA may also prove clinically relevant.18

The primary target for the treatment of RA should be a state of clinical remission.14 The level of evidence supporting this statement is currently low, because strategic trials have previously aimed at attaining low disease activity, while no formal studies have compared a strategy to treat RA to achieve a target of ‘remission’ with other strategies.19

The present study used clinical remission in the form of a disease activity score in 28 joints (DAS28) less than 2.6 as a target for therapy in early RA patients.20 Several studies have shown matrix metalloproteinase (MMP) 3 can be used to predict joint destruction in RA,21 and tocilizumab-treated RA patients who achieved clinical remission showed greater reductions in MMP-3 levels compared with those who did not achieve clinical remission.22 We tested the hypothesis that therapy to achieve both DAS28 less than 2.6 and MMP-3 normalisation (men, <121 ng/ml; women, <59.1 ng/ml)23 as twin targets would achieve better outcomes than therapies targeting DAS28 less than 2.6 alone or MMP-3 normalisation alone in early RA.

Methods

Patients

The T-4 Study was undertaken in two teaching hospitals in Aomori, Japan. Between August 2008 and April 2010, we recruited 243 patients aged between 18 and 75 years who had RA. All patients fulfilled the American College of Rheumatology 1987 criteria for RA.24 Inclusion criteria were as follows: age greater than 18 years; disease duration less than 3 years and DAS28 greater than 3.2. Exclusion criteria were: previous use of glucocorticoids or any disease-modifying antirheumatic drugs (DMARD) or any biological agents; relevant concurrent liver (aspartate aminotransferase >100 IU/l or alkaline phosphatase >100 IU/l), renal (serum creatinine >1.5 mg/dl), haematological (total white blood cell count <4000/ml, platelet count <150 000/ml), or severe respiratory disease; pregnancy, plans to become pregnant, or unwillingness to use effective contraception; presence of hepatitis B, hepatitis C, or HIV; and psychological problems that would make adherence to the study protocol impossible.

Patients were randomly allocated to one of four strategy groups to receive: routine care (R group); DAS28-driven therapy (D group); MMP-3-driven therapy (M group); or both DAS28 and MMP-3-driven therapy (twin; T group).

Treatment

This study used the following visiting times: weeks 0, 2, 4, 6, 8, 12, 16, 20, 24, 28, 32, 36, 40, 44, 48, 52 and 56. At baseline and at predefined assessment points (weeks 0, 2, 4, 8, 12, 16, 24, 32, 40, 48 and 56), the following clinical variables were assessed: erythrocyte sedimentation rate (ESR) (mm/h); number of swollen joints (0–26); number of tender joints (0–28); visual analogue scale for pain (0–100 mm, 100 = most pain); visual analogue scale for general wellbeing (0–100 mm, 100 = worst status); simplified disease activity index (SDAI) and MMP-3 level. ESR was determined locally using the Westergren method. Serum levels of MMP-3 were determined by ELISA (Panaclear MMP-3; Daiichi Fine Chemicals, Takaoka, Japan). Lower values of 121.0 ng/ml (men) and 59.7 ng/ml (women) were used for MMP-3 normalisation categorisation. In the D and T groups, each physician calculated the DAS28 score of their patient on the same day as clinical variables were assessed. Articular examination was performed by the same physician for all patients in the R, D and T groups. The Japanese version of the modified health assessment questionnaire (mHAQ) was completed on every visit (0–3, 3 = most functional disability).25 26

In particular, medication was started with a dose of 1 g/day of sulfasalazine in all intervention groups. In the R group, change of therapy was based on the treating physician's clinical judgement according to the improvement in the number of tender joints (0–28), swollen joints (0–26) and value of serum C-reactive protein (CRP) from preassessment values, without access to current DAS28 and MMP-3 values. Mean target values for each physician were the number of tender joints (0–28) two or less, swollen joints (0–26) two or less and serum CRP 0.7 mg/dl or less. Target values were DAS28 less than 2.6 for the D group, MMP-3 less than 121 ng/ml (men) or less than 59.7 ng/ml (women) for the M group, and both DAS28 less than 2.6 and MMP-3 less than 121 ng/ml (men) or less than 59.7 ng/ml (women) for the T group. If the value in question did not fall below the previously measured level, we added methotrexate. The starting dose of oral methotrexate was 4 mg/week. Folic acid was administered to every patient (5 mg/week) 36 h after methotrexate dosing. In intensive strategy groups, the dosage of methotrexate was not changed if the patient had responded compared with the previous visit; otherwise the dosage was increased in a stepwise manner to a maximum of 8 mg/week if patient had not responded. If the maximum tolerable dose of methotrexate that introduced a dose-dependent side-effect was reached and the patient still did not fulfil the criteria for sustained response, tumour necrosis factor (TNF) blockers were allowed. If patients with the administration of TNF blockers did not show improvement compared with the previous measurement, we changed the TNF blockers to another biological agent, or increased the dose of the TNF blocker, or shortened the interval for TNF administration.

The maximum dosages allowed were 1 g/day for sulfasalazine, 20 mg/day for leflunomide, 300 mg/day for bucillamine, 25 mg/week for gold sodium thiomalate, 3 mg/day for tacrolimus hydrate, 10 mg/kg bimonthly or 6 mg/kg monthly for infliximab, 50 mg/week for etanercept, 40 mg fortnightly for adalimumab and 8 mg/kg monthly for tocilizumab. DMARD including bucillamine, gold sodium thiomalate, tacrolimus and leflunomide were given as allowed by the rheumatologist at all times. Combination therapy with DMARD other than methotrexate was allowed for two kinds of agents. Intra-articular glucocorticoid (to a maximum of 10 mg triamicinolone acetonide on a single visit) was permitted for persistently swollen and tender joints. Patients were given physiotherapy and/or occupational therapy when needed.

Clinical response

Primary outcome measures consisted of the proportion of patients in clinical remission (European League Against Rheumatism definition, DAS28 <2.6). Two radiologists scored radiographs of the hands and feet at baseline and 56 weeks using the van der Heijde modification of the Sharp score (range 0–448), and calculated the change from baseline to 56 weeks in the modified total Sharp score (ΔmTSS).27 Films were scored with the sequence of radiographs known, but radiologists were blinded as to the treatment group. The difference between scores by the two readers for each radiograph was less than 1% of the maximum mTSS score. The secondary outcome measure consisted of the proportion of patients in radiographic non-progression (ΔmTSS ≤0.5), showing normal physical function (mHAQ 0), and in comprehensive disease remission,4 defined as the combination of clinical remission (DAS28 <2.6), radiographic non-progression (ΔmTSS ≤0.5) and normal physical function (mHAQ 0).

Statistical analysis

All available data were used for the analyses. Outcomes for patients between the R, D, M and T groups were compared using Student's t test, the Mann–Whitney U test, Kruskal–Wallis and the χ2 test, as appropriate. For the clinical remission rate, radiographic non-progression rate, normal physical function rate and comprehensive disease remission rate, patients who had to withdraw before completing 56 weeks of the study were designated as non-responders at 56 weeks. Statistical analyses were performed using JMP version 8 software (SAS). All reported p values are two sided and values of p<0.05 were considered significant.

Results

Clinical outcomes

A total of 243 patients with recent-onset, active RA was included (R group, n=62; D group, n=60; M group, n=60; T group, n=61) (figure 1). Baseline characteristics of patients for each strategy group are summarised in table 1. No significant differences in these variables were seen between the four groups. At baseline, 84 of 243 patients (35%) displayed a DAS28 greater than 5.1, indicating severe disease. The positive serum test for rheumatoid factor showed no significant difference between the four groups (R group, 55%; D group, 63%; M group, 43%; T group, 69%).

Figure 1

Randomisation, reasons for discontinuing therapy and numbers of patients who completed the trial. Other reasons for discontinuing therapy included withdrawal of consent and withdrawal because of non-compliance. DAS28, disease activity score in 28 joints; MMP-3 matrix metalloproteinase 3.

Table 1

Baseline characteristics of patients*

During this 56-week study, each therapy was well tolerated. Similar numbers of patients in each group discontinued therapy due to adverse events. Two hundred and twenty-two patients (R group, n=55; D group, n=56; M group, n=53; T group, n=58) were available for analysis of clinical efficacy, radiographic analysis and functional analysis. All 240 patients (R group, n=61; D group, n=59; M group, n=59; and T group, n=61) were included in the safety analysis.

Treatment

Table 2 shows number of patients using concomitant drugs to RA during the T-4 Study. More patients were prescribed TNF blockers in the T group than patients in the other strategy groups (p<0.005). Fewer patients were prescribed glucocorticoids in the T group than patients in the other strategy groups (p<0.05). There were no differences in the rate of methotrexate use and dose of methotrexate between the four groups. At least one intra-articular glucocorticoid was administered to three patients (5%) in the R group, seven patients (12%) in the D group, eight patients (13%) in the M group and five patients (8%) in the T group. The mean frequency of triamicinolone acetonide use did not differ significantly among groups (R group, once; D group, once; M group, three times; T group, once) at 56 weeks. The maximum frequency of triamicinolone acetonide use did not differ significantly among groups (R group, once; D group, once; M group, 13 times; T group, 5 times) at 56 weeks. Table 3 shows the therapeutic interventions and initiations according to the visit in all subgroups.

Table 2

Number of patients using concomitant drugs related to RA during the study (comparing between therapy groups)

Table 3

Population in T-4 groups on therapeutic interventions and initiations according to each visit

Efficacy

Changes in parameters for each therapy group at 56 weeks

No significant change from baseline ESR was seen between the groups at 56 weeks (table 4). CRP levels decreased significantly from baseline in the tight control groups compared with the R group by 56 weeks (p<0.05). No significant differences were apparent among therapy groups in terms of changes in MMP-3 levels at 56 weeks. By 56 weeks, DAS28 levels had decreased significantly from baseline in the D group compared with the R group (p<0.05), but not compared with the other groups. Although a tendency was seen for the T group to show a greater change from baseline than all other groups in the score for mTSS, this difference was only significant compared with the R and D groups. The modified erosion score decreased significantly from baseline in the T group compared with the R group (p<0.01), but not significantly compared with any other groups by 56 weeks. Although the T group tended to show a greater change from baseline compared with all other groups in the score for modified joint-space narrowing, this difference was only significant compared with the R and D groups (p<0.05). No significant change from baseline was identified in the score for mHAQ between therapy groups at 56 weeks.

Table 4

Changes in parameters for each therapy group at 56 weeks*

Patients meeting all criteria for comprehensive disease remission at 56 weeks

Figure 2 shows the proportions of patients who achieved clinical remission (DAS28 <2.6 and SDAI ≤3.3), radiographic non-progression (ΔmTSS ≤0.5), normal physical function (mHAQ 0) and comprehensive disease remission.

Figure 2

Percentage of patients who met triple criteria for comprehensive disease remission and SDAI ≤3.3 at 56 weeks (non-responder imputation). *p<0.05 versus DAS28-driven therapy group and †p<0.0005 versus both DAS28 and MMP-3-driven therapy group. ‡p<0.005 versus MMP-3-driven therapy group. §p<0.0001 versus both DAS28 and MMP-3-driven therapy group. ¶p<0.0005 versus MMP-3-driven therapy group, and **p<0.0005 versus both DAS28 and MMP-3-driven therapy group. ††p<0.05 versus MMP-3-driven therapy group. ‡‡p<0.005 versus both DAS28 and MMP-3-driven therapy group. §§p<0.005 versus both DAS28 and MMP-3-driven therapy group. ¶¶p<0.001 versus both DAS28 and MMP-3-driven therapy group. ***p<0.05 versus both DAS28 and MMP-3-driven therapy group. †††p<0.001 versus both DAS28 and MMP-3-driven therapy group. ‡‡‡p<0.0001 versus both DAS28 and MMP-3-driven therapy group. §§§p<0.05 versus MMP-3-driven therapy group. ¶¶¶p<0.0001 versus MMP-3-driven therapy group. DAS28, disease activity score in 28 joints; mHAQ, modified health assessment questionnaire; MMP-3, matrix metalloproteinase 3; mTSS, modified total Sharp score; SDAI, simplified disease activity index.

Although the T group tended to show a higher proportion of patients achieving clinical remission (DAS28 <2.6) compared with all other groups, this difference was only significant with the R (p<0.0005) and M groups (p<0.0001) at 56 weeks. Although there was a tendency for the M group to show a greater proportion of patients achieving radiographic non-progression (ΔmTSS ≤0.5) than any other group, this difference was only significant with the R (p<0.0005) and D groups (p<0.05) at 56 weeks. Although there was a tendency for the T group to show more patients achieving normal physical function (mHAQ 0) compared with any other group, this difference was only significant with the R (p<0.005) and M groups (p<0.005) at 56 weeks. Significantly more patients in the T group achieved comprehensive disease remission at 56 weeks than in the R group (p<0.001), D group (p<0.05), or M group (p<0.001).

Supplementary figure S1 (available online only) shows the proportions of patients who achieved clinical remission (DAS28 <2.6 and SDAI ≤3.3), radiographic non-progression (ΔmTSS ≤0.5), normal physical function (mHAQ 0) and comprehensive disease remission between moderate (3.2< DAS28 ≤5.1) and high (DAS28 <5.1) disease activity at baseline. Despite high disease activity at baseline, a higher proportion of patients achieved clinical remission (DAS28 <2.6) (53%) in the T group than in the R group (p<0.05) or M group (p<0.05).

Adverse effects

Frequently occurring adverse events identified in patients are summarised in supplementary table S1 (available online only) by strategy groups. The number of serious adverse events did not differ significantly between patients in the four strategy groups. No cases of tuberculosis or demyelinating disease were recognised.

Discussion

This study may represent the tightest control for RA treatment to be described in a study to date. The D group showed superior results to the R group in terms of both clinical remission and the radiographic non-progression rate.

M group adjustments resulted in a significantly better radiographic non-progression rate compared with the R group. However, no significant differences between the D and M groups and the R group were seen for the normal physical function rate. More patients in the T group achieved clinical remission, radiographic non-progression and normal physical function than in the R group. Furthermore, only the T group achieved a moderate level of success in terms of the comprehensive disease remission rate (34%).

Clinical studies providing treatment for the tight control of DAS28 remission targets in RA have not previously been reported.18 We tried to perform treatment with the goal of achieving remission from RA rather than targeting low disease activity. On the other hand, the medications in treatment strategies that have been reported have almost all been methotrexate and DMARD. The only research into treatment strategies including biological agents has been the BeSt Study.13 Unfortunately, clinical remission was not a treatment goal for all strategy groups in the BeSt Study.

The present study including biological agents revealed that if the treatment goal is DAS28 less than 2.6, comprehensive disease remission could be achieved in 15% of patients. However, no significant difference was seen between the R and M groups in terms of the comprehensive disease remission rate. The D group exhibited worse radiographic outcomes, yet the improvement in DAS28 values was even better than in the T4 cohort, because a reduction of DAS28 to less than 2.6 was delayed compared with the T group (24 months (mean value) vs 16 months, p<0.005, respectively). In addition, in the D group, radiographic non-progression rate was significantly inferior to the M and T groups. This result agrees with a previous study that DAS28 is not a predictive factor for radiographic non-progression.28 One contributing reason is that DAS28 does not include joints of the feet, which are frequently inflamed in RA.29 A target of DAS28 less than 2.6 may thus be a suboptimal target for preventing structural damage in early RA.

MMP-3, a proteinase that is expressed in rheumatoid synovial tissue and shows potent activity in degrading proteoglycans in cartilage, plays a pivotal role in the joint destruction seen in early RA. Serum concentrations of MMP-3 offer a useful marker for predicting bone damage in the early stage of RA,30 and suppression of MMP-3 production may be an effective therapeutic approach for patients with early RA. The MMP-3 target tended to be more challenging than the DAS28 target. More patients in the M group had achieved radiographic non-progression (62%) by 56 weeks, but fewer patients displayed clinical remission and normal physical function. The comprehensive disease remission rate in the M group showed no significant difference compared with the R group. MMP-3 alone was not found to be a predictive factor for comprehensive disease remission. However, this result might change if MMP-3 could be normalised earlier.

This study does have some limitations that need to be considered, including the open approach and small sample size. Methotrexate is a highly effective drug for disease modification in RA, and more recent insights suggest that methotrexate at higher weekly doses (20–30 mg) is more effective than methotrexate at lower weekly doses (7.5–15 mg).2 31 It should be noted that methotrexate was used at doses relatively lower than the optimal doses currently recommended in a recent meta-analysis, with treatment started at 15 mg/week of oral methotrexate. However, in Japan, the central government has approved a maximum dose of 8 mg/week for methotrexate in RA patients as of February 2011, and access to subsidised methotrexate therapy is limited to patients displaying previous failure on DMARD. The combination of high-dose methotrexate and biological agents could deliver better results. Moreover, in Japan, the approach with infliximab is unique. Normally, infliximab is administered intravenously via drip infusion at a dose of 3 mg/kg bodyweight, 2 weeks and 6 weeks after the initial dose and thereafter every 8 weeks. If effects are insufficient or have diminished following administration at 6 weeks, the dosage may be increased or the administration interval shortened. These increases in dosage and shortening of the administration interval are to be carried out in incremental steps. If the administration interval is 8 weeks the upper limit for a dose per kilogram bodyweight is 10 mg, and if the administration interval has been shortened the upper limit for a dose is 6 mg. The shortest administration interval is 4 weeks. However, this study protocol affected ‘real world care for RA patients in Japan’.

The effects of these intensive therapy groups on long-term progression in RA patients are unknown. Longer follow-up studies of RA patients with these four strategies are therefore needed and more data on clinical, radiographic and functional outcomes among patients with established RA are needed.

We hypothesised that twin targets (both MMP-3 and DAS28) will yield better results than either MMP-3 alone or DAS alone as targets for the treatment of RA patients. DAS28 is inadequate as a prognostic factor for radiographic non-progressive status and MMP-3 is thought to supplement this unsatisfactory index. The group receiving twin therapy showed the best results in terms of the clinical remission rate, radiographic non-progressive rate, normal physical function rate and comprehensive disease remission rate among the four strategy groups. A single target strategy cannot achieve high rates of comprehensive disease remission rate.

Acknowledgments

The authors would like to thank all the patients and research personnel for their assistance in collecting patient data. They dedicate this article to the patients who suffered in the Great Eastern Japan Earthquake.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

    • Web Only Data - This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
    • Web Only Data - This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval The study protocol was approved by the institutional review committee at each participating centre.

  • Provenance and peer review Not commissioned; externally peer reviewed.