5-year remission rate after the discontinuation of adalimumab in patients with rheumatoid arthritis: Long-term follow-up results of the HONOR study

Abstract

Objectives: To determine the rate and factors associated with remission (disease activity score (DAS) 28-erythrocyte sedimentation rate (ESR) of <2.6) during a 5-year follow-up after the discontinuation of adalimumab (ADA) in patients with rheumatoid arthritis (RA).

Methods: 75 patients who had been treated with ADA + methotrexate (MTX) and maintained DAS28-ESR <2.6 for at least 6 months were enrolled. Among them, 52 patients discontinued ADA, and 46 patients completed a 5-year follow-up.

Results: During the 5 years, 11 patients had DAS28-ESR <2.6. In 15 patients with DAS28-ESR <3.2, no significant changes were found in the health assessment questionnaire disability index (HAQ-DI) and modified total Sharp score (mTSS). When comparing patients with DAS28-ESR ≤1.61 versus 1.61 <DAS28-ESR <2.6, 50% and 15% of the two groups demonstrated sustained remission, respectively. Remission was more common in patients with shorter disease duration (≤2 years) than those with longer duration (>2 years). Among 31 patients who experienced flare, ADA was restarted in 24 patients, and 17 patients of these achieved DAS28-ESR <3.2 within 1-year.

Conclusion: During the 5-year ADA-free period, remission rate was persistent in 21% of the patients. ADA-free remission was possible especially in patients with deeper remission (DAS28-ESR ≤1.61) and shorter disease duration (≤2 years).

Keywords:

• Discontinuation
• HONOR study
• remission
• rheumatoid arthritis
• TNF inhibitor

Introduction

Rheumatoid arthritis (RA) is a progressive inflammatory disease that can cause joint destruction and functional disability [1,2]. The introduction of biologics, such as tumor necrosis factor (TNF) inhibitors, in the treatment of RA has increased clinical remission rates and enabled long-term maintenance of structural and functional remission [3,4]. Maintaining all patients with RA in remission without any medications should be the next goal in the treatment of RA [5,6]. In fact, several studies had reported that at least some patients with RA can go into remission for 1–2 years without any medications. In this regard, we have demonstrated that the maintenance of remission is possible in RA, especially in patients with RA who had received early treatment and in whom disease activity was well controlled before discontinuation of adalimumab (ADA), an anti-TNF agent [7,8]. However, little has been reported on the long-term outcome of biologics-free remission [9–13]. Since RA is a chronic disease, it is also important clinically to determine the long-term consequences of discontinuation of biologics. The Humira discontinuation withOut functional and radiographic damage progressioN follOwing sustained Remission (HONOR) study was conducted to elucidate the clinical, functional and radiographic outcomes after discontinuation of ADA. As an extension to our two previous studies [7,8], we conducted the present study 5 years after the above two studies to determine the long-term outcome after discontinuation of ADA. Specifically, we followed the patients with RA clinically, including laboratory functional tests as well as the van der Heijde’s modified total Sharp score (mTSS).

Methods

Study design

This study was based on the HONOR study, an open-label, non-randomized trial that was approved by the ethics review board of the University of Occupational and Environmental Health Japan, and registered at the University Hospital Medical Information Network-clinical Trials Registry (UMIN-CTR) as UMIN000006669. The study was designed to evaluate disease activity, functional disability and progression of radiographically documented articular damage after discontinuation of ADA. Patients received subcutaneous injection of 40 mg ADA combined with methotrexate (MTX) every other week. Administration of biologics and disease-modifying anti-rheumatic drugs (DMARDs) and oral steroids was at the discretion of the attending rheumatologist, but treatment with ADA + MTX was initiated with the aim of induction of remission on the same patients whenever appropriate. The decision to discontinue ADA was based on the physician’s judgment and patients’ agreement. The criteria for the selection of ADA-free course were as follows. (1) Persistent stable remission (DAS28-ESR <2.6) [14] with ADA + MTX for at least 6 months. (2) No treatment with glucocorticoids, non-steroidal anti-inflammatory drugs (NSAIDs), or cyclooxygenase-2 (COX-2) inhibitors. (3) Use of a stable dose of MTX for at least 12 weeks [15]. Patients who had been maintained at DAS28-ESR <2.6 with ADA + MTX were enrolled in the discontinuation study after providing a signed informed consent. Patients with flare, defined as DAS28-ESR ≥3.2, were rescued by re-administration of ADA or other treatments, such as increases in the dose of MTX. We collected demographic information (age, gender, disease duration), duration of treatment with ADA, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), disease activity score (DAS) 28-ESR, clinical disease activity index (CDAI), simplified disease activity index (SDAI), health assessment questionnaire disability index (HAQ-DI), rheumatoid factor (RF), level of matrix metalloproteinase-3 (MMP-3), dose of MTX, and results of imaging using the mTSS. The study was conducted in compliance with the Helsinki Declaration.

Patients

The study subjects were male and female adult patients (age >18 years) who met the 1987 American College of Rheumatology (ACR) criteria [16] for RA and showed inadequate clinical response to MTX (4–16 mg/weeks according to the Japanese MTX package insert) and/or had other non-DMARDs initiated ADA treatment between July 2008 and April 2011, according to the Japanese package insert or the Japan College of Rheumatology (JCR) for anti-TNF drugs. As stated above, all the treatment decisions taken throughout the study were based on the JCR guidelines and judgment by the attending rheumatologists and selection by the patients. Of the 197 patients who started ADA (40 mg subcutaneously every other week) combined with MTX (mean dose: 8.8 ± 3.0 mg/week, ±standard deviation (SD)) under the Japanese National Insurance System between July 2008 and April 2011, 75 patients met the criteria for discontinuation: achievement and maintenance of DAS28-ESR <2.6 for at least 24 weeks (Figure 1). However, instead of discontinuation by all patients, 23 patients continued to receive ADA for at least another 24 weeks (maximum 1 year) with DAS28-ESR <2.6 throughout this period, and served as the control group. The remaining 52 patients entered the HONOR study with discontinuation of ADA. Of these 52 patients, 46 patients completed the follow-up study period of 5 years (260 weeks), while six patients withdrew because they moved after 52 weeks (Figure 1).

Figure 1. Patient recruitment process. ADA: adalimumab; HONOR: Humira discontinuation withOut functional and radiographic damage progressioN follOwing sustained Remission; DAS28: disease activity score 28; ESR: erythrocyte sedimentation rate.

Outcome measures

The primary outcome measure was the percentage of patients who maintained remission with DAS28-ESR <2.6 at 5 years after discontinuation of ADA. The secondary outcome measures included the percentages of patients who maintained clinical remission based on CDAI (≤2.8) or SDAI (≤3.3), the percentage of patients who maintained low disease activity (LDA) clinically according to CDAI (≤10), SDAI (≤11), and DAS28-ESR (<3.2) at 5 years after discontinuation of ADA. The functional and radiographic effects of treatment discontinuation were examined by the HAQ-DI [17] and mTSS [18]. Further subanalysis was conducted according to disease duration, stratified at 2 years, and the need for re-administration of ADA.

Statistical analysis

Continuous data were expressed as mean ± SD, while categorical data as number (%). Differences between groups in patients’ baseline demographic and clinical characteristics were analyzed by t-test or the Wilcoxon rank sum test as appropriate for continuous variables while Fisher’s exact test was used for comparison of proportions. A survival curve was constructed using the Kaplan–Meier estimates. The odds ratios (ORs) were calculated to estimate the relationship between sustained remission and the variables mentioned above. Multivariate logistic regression analysis was conducted using covariates selected based on clinical relevance (age, disease duration, duration of ADA use, DAS28-ESR, HAQ-DI, RF, MMP-3, MTX doses, mTSS) to identify those factors related to sustained remission over 5 years after discontinuation of ADA. A receiver operating characteristic (ROC) curve analysis was conducted to determine the cut-off points of various variables at study entry. In all analyses, a p value of <.05 was considered to denote significant difference. The last observation carried forward (LOCF) was used for missing clinical or functional values after the start of ADA discontinuation. All analyses were performed using JMP version 12.2 (SAS Institute Inc., Cary, NC) or Prism version 7.0d (Graph Pad Software Inc., San Diego, CA).

Results

Changes at termination of ADA

All 52 patients with RA fulfilled the ADA-free criteria and agreed on discontinuation of ADA. They were treated with ADA from July 2008 to April 2011 (34 months). They included 11 men and 41 women with a mean age of 59.8 years and mean disease duration of 7.0 years. Table 1 shows changes in initiation and termination of ADA.

Table 1. Baseline characteristics of patients with RA at initiation and termination of ADA.

Measurement items	Initiation of ADA (n = 52)	Termination of ADA (n = 52)
Age (years)	59.8 ± 11.3	61.2 ± 11.5
Disease duration (years)	7.0 ± 9.8	8.5 ± 9.8
ESR (mm/h)	43.0 ± 33.3	15.5 ± 10.1
CRP (mg/dL)	2.2 ± 3.7	0.1 ± 0.2
DAS28-ESR	5.1 ± 1.2	1.9 ± 0.5
CDAI	21.4 ± 12.9	1.0 ± 1.3
SDAI	23.4 ± 13.9	1.1 ± 1.3
HAQ-DI (0–3)	0.9 ± 0.7	0.2 ± 0.3
RF (U/mL)	108.1 ± 141.8	43.7 ± 54.2
MMP-3 (mg/mL)	226.6 ± 343.0	52.6 ± 27.4
MTX (mg/W)	9.0 ± 2.7	8.5 ± 2.2
mTSS	43.1 ± 83.4	42.2 ± 83.6

ADA: adalimumab; ESR: erythrocyte sedimentation rate; CRP: C-reactive protein; DAS28: disease activity score 28; CDAI: clinical disease activity index; SDAI: simplified disease activity index; HAQ-DI: Health Assessment Questionnaire-Disability Index; RF: rheumatoid factor; MMP-3: matrix metalloproteinase-3; MTX: methotrexate; mTSS: modified total Sharp score; SD: standard deviation.

Data are mean ± SD.

Clinical disease activity

Among the 52 study patients, 46 patients completed the 5-year follow-up, six patients dropped out because they moved (Figure 1). Figure 2 shows the time course of DAS28-ESR in patients for 5 years after ADA discontinuation. Of these, remission was noted in 11/52 (21%) as defined by DAS28-ESR <2.6, 15/52 (29%) based on DAS28-ESR <3.2 (Figure 2(A)) as the final result. The maintained DAS28-ESR remission was estimated 21% in the Kaplan–Meier method (Figure 2(B)). 13/52 patients as defined by CDAI ≤2.8, 19/52 patients based on CDAI ≤10, 13/52 patients as defined by SDAI ≤3.3, and 19/52 patients based on SDAI ≤11.

Figure 2. 5-year remission (DAS28-ESR <2.6) rate after discontinuation of ADA. (A) Clinical outcome was evaluated by DAS28-ESR. Changes in DAS28-ESR and proportions of restart of ADA after discontinuation of ADA. (B) The remission estimate based on the maintained DAS28-ESR <2.6 for 5 years was analyzed by the Kaplan–Meier survival curve. DAS28: disease activity score 28; ESR: erythrocyte sedimentation rate; ADA: adalimumab.

Effects of discontinuation of ADA in 15 patients with LDA

15 patients sustained LDA (DAS28-ESR <3.2) for 5 years after discontinuation of ADA. In these patients, no significant changes were found in HAQ-DI (p = .483) and mTSS (p=.475) during the discontinuation of ADA (Figure 3). Structural remission (ΔmTSS ≤0.5/year) was noted in 8/9 patients as the final result. One patient who originally had a high mTSS score did not have structural remission. The above results suggest that maintenance of LDA facilitates functional and structural remission for at least 5 years after discontinuation of ADA. In 15 patients with LDA, the dose of MTX was reduced from 8.23 ± 2.62 to 6.32 ± 4.23 mg/week within 5 years during the discontinuation of ADA (p=.091). Of the group, the dose of MTX was reduced in 6 patients and stopped entirely in 2 patients after 5 years.

Figure 3. Values of HAQ-DI and mTSS score in 15 patients with sustained DAS28-ESR <3.2 after the discontinuation of ADA. Data of patients who showed sustained DAS28-ESR <3.2 without flaring for 5 years after the discontinuation of ADA. (A, B) Values of HAQ-DI and mTSS score. Statistical significance was assessed by the t-test (p<.05). HAQ-DI: health assessment questionnaire-disability index; mTSS: modified total Sharp score; DAS28: disease activity score 28; ESR: erythrocyte sedimentation rate; ADA: adalimumab.

Maintenance of DAS28-ESR at <2.6 for 5 years after the discontinuation of ADA

Comparison of patient background at discontinuation of ADA between those who experienced sustained DAS28-ESR <2.6 for 5 years (n = 11) and those that did not (n = 35) showed significant differences in four items: female (54.6 vs. 85.7%, p=.043), RA disease duration (4.5 vs. 10.1 years, p=.023), ESR (10.4 vs. 17.3 mm/h, p=.032), and DAS28-ESR score (1.6 vs. 2.0, p=.030) (Table 2).

Table 2. Characteristics of patients who sustained or did not sustain remission (DAS28-ESR <2.6) for 5 years after the discontinuation of ADA.

Measurement items	Sustained DAS28-ESR <2.6 (n = 11)	Failed DAS28-ESR <2.6 (n = 35)	p Value
Age (years)	58.8 ± 13.7	61.6 ± 9.9	.553
Gender, female (%)	54.6	85.7	.043*
Disease duration (years)	4.5 ± 7.2	10.1 ± 10.6	.023*
ADA admin periods (weeks)	65.0 ± 28.0	69.3 ± 29.2	.521
ESR (mm/h)	10.4 ± 6.3	17.3 ± 10.3	.032*
CRP (mg/dL)	0.1 ± 0.2	0.1 ± 0.2	.150
DAS28-ESR	1.6 ± 0.5	2.0 ± 0.4	.030*
CDAI	0.8 ± 1.0	1.1 ± 1.4	.897
SDAI	0.9 ± 0.9	1.2 ± 1.5	.969
HAQ-DI (0–3)	0.09 ± 0.2	0.3 ± 0.3	.122
RF (U/mL)	45.8 ± 45.3	43.4 ± 60.2	.757
MMP-3 (mg/mL)	59.0 ± 26.6	49.5 ± 28.4	.110
MTX (mg/W)	8.0 ± 3.0	8.7 ± 1.9	.528
mTSS	13.5 ± 22.0	53.5 ± 99.0	.500

ESR: erythrocyte sedimentation rate; CRP: C-reactive protein; DAS28: disease activity score 28; CDAI: clinical disease activity index; SDAI: simplified disease activity index; HAQ-DI: Health Assessment Questionnaire-Disability Index; RF: rheumatoid factor; MMP-3: matrix metalloproteinase-3; MTX: methotrexate; mTSS: modified total Sharp score; ADA: adalimumab; SD: standard deviation.

Data are mean ± SD.

* By Wilcoxon’s rank sum test (p<.05).

Prognostic factors affecting 5-year remission at discontinuation of ADA

Next, we analyzed the data for factors that could predict sustained remission for 5 years. The following parameters were included: age, disease duration, duration of administration of ADA, DAS28-ESR, HAQ-DI, RF, MMP-3, MTX doses and mTSS at discontinuation of ADA. The results of multivariate logistic regression analysis identified DAS28-ESR at the time of discontinuation to correlate significantly and independently with sustained remission in logistic (OR, 0.017; 95% confidence interval (95%CI), 0.000–0.588; p=.024) (Table 3). Furthermore, ROC curve analysis for estimation of sustained remission found a cut-off value of 1.61 for DAS28-ESR, with sensitivity of 55%, specificity 83% and area under curve (AUC) of 0.72. These results highlight the importance of remission before discontinuation of ADA in patients with established RA.

Table 3. Multivariate logistic regression analysis of the prognostic factors for discontinuation of ADA based on sustained remission (DAS28-ESR < 2.6) at 5 years.

Model	Odds ratio	95%CI	p Value
Age (years)	0.907	0.790–1.041	.163
Disease duration (years)	1.111	0.852–1.449	.437
ADA admin periods (weeks)	6.865	0.464–101.505	.161
DAS28-ESR	0.017	0.000–0.588	.024*
HAQ-DI (0–3)	0.019	0.000–3.355	.133
RF (U/mL)	0.989	0.971–1.007	.214
MMP-3 (mg/mL)	1.023	0.984–1.063	.251
MTX (mg/W)	0.545	0.294–1.012	.054
mTSS	0.971	0.919–1.026	.298

CI: confidence interval; ESR: erythrocyte sedimentation rate; DAS28: disease activity score 28; HAQ-DI: Health Assessment Questionnaire-Disability Index; RF: rheumatoid factor; MMP-3: matrix metalloproteinase-3; MTX: methotrexate; mTSS: modified total Sharp score; ADA: adalimumab.

Multivariate analyses included covariates measured at discontinuation of ADA to determine those factors related to sustained remission for 5 years after discontinuation of ADA.

* A significant difference (p<.05): Wald’s test.

Comparison between partial and complete remission

We defined complete remission as DAS28-ESR below 1.61 and partial remission as DAS28-ESR ranging from 1.61 to 2.6. Baseline characteristics of the complete remission group (n = 12) and the partial remission group (n = 34) are presented in Table 4. The complete group had fewer females (41.7%). Comparison of patients of the complete remission group with those of the partial remission showed sustained remission (DAS28-ESR <2.6) in 50% vs. 14.7%, respectively, during the discontinuation of ADA (p=.022). Furthermore, the proportion of patients who maintained LDA (DAS28-ESR <3.2) was 50% vs. 26.5% (p=.165), respectively (Figure 4).

Figure 4. Effect of extent of clinical remission during the ADA-free follow-up. Percentages of patients at remission and low disease activity (LDA) at 5 years after the adalimumab (ADA) discontinuation in patients with complete and partial remission, based on a cut-off value of disease activity score (DAS) 28-erythrocyte sedimentation rate (ESR) of ≤1.61 on receiver operating characteristics (ROC) analysis. Statistical significance was assessed by Fisher’s exact test (p<.05).

Table 4. Baseline characteristics before treatment with ADA between patients who were included in the study with a DAS28-ESR ≤1.61 and 1.61 < DAS28-ESR <2.6.

Measurement items	DAS28-ESR ≤1.61 (n = 12)	1.61 < DAS28-ESR <2.6 (n = 34)
Age (years)	58.8 ± 11.2	59.7 ± 10.7
Gender, female (%)	41.7	91.2
Disease duration (years)	7.4 ± 9.9	7.3 ± 10.3
ESR (mm/h)	39.3 ± 35.7	44.6 ± 33.7
CRP (mg/dL)	2.3 ± 0.9	1.9 ± 0.5
DAS28-ESR	4.9 ± 1.5	5.1 ± 1.3
CDAI	22.1 ± 15.3	21.2 ± 12.9
SDAI	24.4 ± 15.5	22.9 ± 14.1
HAQ-DI (0–3)	1.1 ± 0.8	0.9 ± 0.6
RF (U/mL)	173.8 ± 178.2	95.6 ± 133.4
MMP-3 (mg/mL)	205.8 ± 173.8	229.0 ± 402.2
MTX (mg/W)	7.6 ± 3.1	9.5 ± 2.2
mTSS	47.8 ± 94.1	42.8 ± 85.9

ESR: erythrocyte sedimentation rate; CRP: C-reactive protein; DAS28: disease activity score 28; CDAI: clinical disease activity index; SDAI: simplified disease activity index; HAQ-DI: Health Assessment Questionnaire-Disability Index; RF: rheumatoid factor; MMP-3: matrix metalloproteinase-3; MTX: methotrexate; mTSS: modified total Sharp score; ADA: adalimumab; SD: standard deviation.

Data are mean ± SD.

Comparative analysis by disease duration

We previously had a report of the sub-analysis of the HONOR study comparing 1-year outcome of DAS28-ESR with disease duration stratified at 2 years [19]. Herein, we also stratified the patients treated with the combination of ADA and MTX based on disease duration of RA, into short-disease duration group (≤2 years, n = 23) and long-disease duration groups (>2 years, n = 23). Baseline characteristics of short-disease duration group and long-disease duration groups are presented in Table 5. Sustained remission, as reflected by the DAS28-ESR, was noted in both groups, but LDA (DAS28-ESR <3.2) was significantly higher in the short-disease duration group, with remission (DAS28-ESR <2.6) rate of 43.5% vs. 4.4% (p=.004) and LDA rate of 56.5% vs. 8.7% (p=.001), respectively (Figure 5).

Figure 5. Effect of RA disease duration on outcome of ADA-free follow-up. Percentages of patients at remission and low disease activity (LDA) at 5 years after adalimumab (ADA) discontinuation in patients with early (disease duration ≤2 years) and established (disease duration >2 years) rheumatoid arthritis (RA). DAS28: disease activity score 28; ESR: erythrocyte sedimentation rate; ADA: adalimumab.

Table 5. Baseline characteristics before treatment with ADA to compare disease duration ≤2 years and >2 years.

Measurement items	Disease duration ≤2 years (n = 23)	Disease duration >2 years (n = 23)
Age (years)	58.9 ± 12.1	60.1 ± 9.3
Gender, female (%)	69.6	87.0
Disease duration (years)	0.9 ± 0.5	13.7 ± 11.2
ESR (mm/h)	48.7 ± 41.6	37.7 ± 23.7
CRP (mg/dL)	2.7 ± 4.1	1.3 ± 1.5
DAS28-ESR	5.1 ± 1.4	5.0 ± 1.2
CDAI	22.4 ± 15.1	20.5 ± 11.6
SDAI	24.9 ± 16.1	21.8 ± 12.5
HAQ-DI (0–3)	1.0 ± 0.8	0.8 ± 0.6
RF (U/mL)	139.9 ± 167.4	92.0 ± 125.8
MMP-3 (mg/mL)	230.0 ± 378.3	215.9 ± 338.7
MTX (mg/W)	8.7 ± 2.8	9.3 ± 2.3
mTSS	7.0 ± 7.3	89.0 ± 115.2

ESR: erythrocyte sedimentation rate; CRP: C-reactive protein; DAS28: disease activity score 28; CDAI: clinical disease activity index; SDAI: simplified disease activity index; HAQ-DI: Health Assessment Questionnaire-Disability Index; RF: rheumatoid factor; MMP-3: matrix metalloproteinase-3; MTX: methotrexate; mTSS: modified total Sharp score; ADA: adalimumab; SD: standard deviation.

Data are mean ± SD.

Effects of re-administration of ADA

During the 5-year ADA-free period, 31/52 patients (60%) experienced flaring (DAS28 ≥ 3.2). In 31 patients with DAS28-ESR ≥3.2 who experienced temporary clinical deterioration, the dose of MTX was significantly increased from 8.69 ± 1.99 to 10.68 ± 3.98 mg/week (p=.012). Of these, 24/52 patients (46%) required a re-start of treatment with ADA, based on the opinion of the attending physician. After the restart of ADA, 15/24 and 17/24 patients achieved remission (DAS28-ESR <2.6) and LDA (DAS28-ESR <3.2) within 1 year, respectively. The median time period between discontinuation to the restart of ADA was 1.3 years (68 weeks). Further analysis showed that patients re-sustained LDA after ADA re-administration in a larger proportion of patients who experienced flaring less than 1 year of discontinuation of ADA (9/10 patients) compared to those who experienced flaring more than 1 year of discontinuation of ADA (8/14 patients) (p=.172).

Safety

Table 6 summarizes the adverse events reported by patients of the HONOR study. No serious adverse events or related deaths were encountered during the study period. The rate of infections was higher in the administration of ADA period compared with the discontinuation of ADA period. Infections were identified in 9 patients of the administration of ADA period, with herpes zoster in four, bronchitis in one, pneumonia in one, urinary tract infection in two, and candidiasis in one. On the other hand, infections were identified in 3 patients of the discontinuation of ADA period, including herpes zoster in two and bronchitis in one. Thus, discontinuation of ADA seems to have reduced the rate of infections.

Table 6. Summary of adverse events experienced by patients in the HONOR study.

Parameters	Administration period of ADA (n = 52)	Discontinuation period of ADA (n = 52)
Total AE	10 (19.2%)	7 (13.5%)
Serious AE	0 (0.0%)	0 (0.0%)
Herpes zoster	4 (7.7%)	2 (3.8%)
Bronchiolitis	1 (1.9%)	1 (1.9%)
Pneumonia	1 (1.9%)	0 (0.0%)
Urinary tract infection	2 (3.8%)	0 (0.0%)
Oral candida	1 (1.9%)	0 (0.0%)
Lymphoproliferative disorders	1 (1.9%)	0 (0.0%)
Malignant tumor	0 (0.0%)	2 (3.8%)
Connective tissue disorders	0 (0.0%)	2 (3.8%)

AE: adverse event; ADA: adalimumab.

Discussion

The use of MTX and biologics in the treatment of RA induces sustained clinical remission. Can such remission be sustained after the discontinuation of all treatments? Several recent studies in fact have reported that at least a proportion of patients with RA can go into remission for 1–2 years without any treatment [20–22]. However, there is less evidence of a longer remission period. The HONOR study is one of a few studies that was designed to establish the effect of discontinuation of ADA on sustained remission (DAS28-ESR < 2.6) in patients with RA. We have already published the 1-year outcome after discontinuation of ADA [8] and the results demonstrated that 48% of patients who met the ADA-free criteria showed sustained DAS28-ESR-based remission after the discontinuation of ADA. The same study also demonstrated that among the patients who maintained DAS28-ESR at <3.2 during the discontinuation of ADA, all continued to show structural remission (ΔmTSS ≤0.5/year) and 94% showed persistent functional remission (HAQ-DI ≤0.5). The present study showed that 21% of the patients were able to maintain DAS28-ESR remission at the extended follow-up period of 5 years after the discontinuation of ADA. Furthermore, about half of the patients with DAS28-ESR score below 1.61 showed sustained DAS28-ESR-based remission. Interestingly, a low score of DAS28-ESR was identified in the HONOR study as a significant factor for 5-year biologics-free remission, and the same study reported that the cut-off values of DAS28-ESR for 6-month- [7], 1-year- [8] and 5-year-sustained remission were 2.16, 1.98 and 1.61, respectively. These results demonstrated that the lower DAS28-ESR score at the time of discontinuation, the longer the biologics-free remission period. Several studies have demonstrated that the use of ADA is followed by long medication-free remission in patients with early RA (disease duration <2 years). In the Japanese HOPEFUL-3 study [12], 65% of patients with sustained clinical remission (DAS28-ESR <2.6) in the discontinuation of ADA group for 3 years. In the Dutch IMPROVED study [13], 12 of 59 (20%) patients who switched to MTX plus ADA (arm2) achieved drug-free remission (DAS44 < 1.6) for 5 years. In the HONOR study, 10 of 23 (44%) patients who started ADA treatment within 2 years of onset achieved ADA-free remission.

Discontinuation of biologic DMARDs preparation after remission is an important issue with respect to safety and cost reduction. After 5 years of follow-up, 15 patients maintained DAS28-ESR at <3.2 while being treated with a smaller MTX dosage (6.32 ± 4.23 mg/week) during the discontinuation of ADA. Of these, dose-reduction was achieved in 6 patients and 2 patients were MTX-free after 5 years. Patients who achieved remission with ADA were able to maintain the remission even after stopping ADA with a small dose of MTX. In addition, the adverse events of infection were rare in those patients who stopped ADA.

In conclusion, we have demonstrated in the present study that ADA ‘treatment holiday’ was feasible for 5 years in at least some patients with established RA who had undergone 2-year treatment from disease onset with the combination of ADA and MTX, without steroids, and showed complete remission during a period of 6 months. Our results also showed that 5-year ADA ‘treatment holiday’ was feasible in about half of the patients who had been diagnosed earlier with RA and received the above combination treatment within 2 years of onset and showed remission. Lastly, ADA ‘treatment holiday’ was feasible for certain period of time; almost of patients who re-started ADA within 1-year showed sustained REM or LDA again. For patients at low risk of flaring, discontinuation of ADA is not associated with worsening of disease activity but rather reduces costs of clinical management and associated adverse effects.

Ethical approval

Ethics review board of the University of Occupational and Environmental Health, Japan.

Author contributions

SH, KS and YT contributed to the conception and design of the study. All authors enrolled and managed the patients in clinic. SH, SK, SF and KH participated in radiographic evaluation. SH and AY performed the statistical analysis. All authors read and approved the final manuscript.

Conflict of interest

YT has received speaking fees and/or honoraria from Daiichi-Sankyo, Astellas, Chugai, Eli Lilly, Pfizer, Abbvie, YL Biologics, Bristol-Myers, Takeda, Mitsubishi-Tanabe, Novartis, Eisai, Janssen, Teijin and has received research grants from Asahi-kasei, Mitsubishi-Tanabe, Chugai, Takeda, Sanofi, Bristol-Myers, UCB, Daiichi-Sankyo, Eisai, Ono. SH has received consultancy fee, advisory fee, or speakers bureau from Abbvie, Asahi-Kasei Pharma, Asahi-Kasei Medical, Astellas, Ayumi, Bristol-Myers Squibb, Celgene, Chugai, Eisai, Eli Lilly, Janssen, Kissei, Novartis, Pfizer, Sanofi, Takeda, Tanabe-Mitsubishi, and UCB. KS has received speaking fees from Eli Lilly. SN has received speaking fees from Bristol-Myers, Sanofi, Abbvie, Eisai, Chugai, Pfizer, Takeda, Asahi-kasei and also research grants from Mitsubishi-Tanabe, Novartis and MSD. KN has received speaking fees from Astellas, UCB, Mitsubishi-Tanabe, Eisai and has received research grants from Mitsubishi-Tanabe, Eisai, and Eli Lilly. SK has received speaking fees from Bristol-Myers, Pfizer, Takeda, and Eli Lilly. The other authors declare no conflict of interest.

Acknowledgements

The authors thank all the medical staff at all participating institutions for providing the data.