Long-term outcomes after discontinuing biological drugs and tofacitinib in patients with rheumatoid arthritis: A prospective cohort study.

OBJECTIVE: This study examined long-term outcomes of biological disease-modifying antirheumatic drugs (bDMARDs) and tofacitinib discontinuation in patients with rheumatoid arthritis (RA).
METHODS: Ninety-seven RA patients who desired drug discontinuation after sustained remission or low disease activity for at least 48 weeks due to stable treatment with biological drugs or tofacitinib were enrolled into this study. All patients were prospectively followed until disease flare or the end of the study. Discontinued drugs (previous drugs) were reintroduced to treat flares.
RESULTS: Following bDMARD/tofacitinib discontinuation (mean follow-up, 2.1 years; standard deviation, 2.0), disease flare occurred at a crude incidence rate of 0.36 per person-year. The median time to flare was 1.6 years (95% confidence interval [CI] 0.9-2.6), and the cumulative flare probability was estimated to be 45% at 1 year, 64% at 3 years, and 80% at 5 years. No or little radiological progression was shown in 87.1% of patients who maintained remission for 3 years. A Fine‒Gray competing risk regression analysis showed that predictive factors for a flare were longer RA duration at the start of bDMARD/tofacitinib treatment, previous failure of treatment with bDMARDs, and a shorter period of remission or low disease activity before drug discontinuation. Type of discontinued drug was not identified as a predictive factor after adjusting for other predictor variables. Restarting previous treatment regimens led to rapidly regaining disease control in 89% of flare patients within 1 month.
CONCLUSION: Discontinuation of bDMARD/tofacitinib may be a feasible strategy in RA patients, especially patients with early treated and longer-controlled RA. Flares are manageable in most RA patients and radiological progression is rare for at least 3 years in patients with sustained remission after bDMARD/tofacitinib discontinuation.

Introduction

Rheumatoid arthritis (RA) is a chronic immune-mediated inflammatory disease that causes progressive damage to joint cartilage and bone, disability, and comorbidity [1]. RA has become a controllable disease. Long-term health-related patient quality of life has dramatically improved because of the availability of conventional, biological, and new non-biological targeted disease-modifying antirheumatic drugs (DMARDs) as well as application of early aggressive intervention and the treat-to-target approach with tight control of disease activity [2]. Today, clinical remission early in the disease course, with a low disease activity (LDA) as a best feasible alternative, is the therapeutic target for every RA patient. Maintaining a state of remission or LDA will relieve clinical signs and symptoms, inhibit the occurrence or progression of structural damage, and improve or normalize physical function and social and work-related activities [3-5]. However, adverse events and high healthcare costs associated with life-long immune-modulatory treatment are significant burdens on patients as well as rheumatologists. Patients frequently express a desire to decrease the dose and for drug holidays [6, 7].

Many studies have addressed the impact of tapering and/or discontinuation of biological DMARDs (bDMARDs) on clinical outcomes in RA patients who have achieved and maintained remission or LDA, including prospective observational studies, randomized controlled trials (RCTs), and open-label extension studies [8-12]. Outcomes of dose reduction and/or discontinuation of targeted synthetic DMARDs (tsDMARDs) have also been reported [13-15]. These studies suggested that a subset of patients may maintain remission or LDA after tapering and even discontinuation of bDMARDs and tsDMARDs. In most studies, however, follow-up periods after tapering and discontinuation of drugs were within 1 year.

In this observational study, we examined the long-term outcomes of bDMARD and tsDMARD discontinuation in RA patients who had maintained remission or LDA for ≥48 weeks. The cumulative probability of disease flare, predictive factors, radiological progression rate, and rescue therapy results were examined.

Patients and methods

Patients

Participants in this study were patients with RA who desired drug discontinuation after maintaining remission or LDA for ≥48 weeks through stable drug treatment with bDMARDs or tsDMARDs from January 2007 to June 2020 in the Rheumatology department of the National Hospital Organization Kumamoto Saishun Medical Center. All participants were over 18 years of age at enrollment and fulfilled the 1987 American College of Rheumatology (ACR) criteria or the 2010 ACR/European League Against Rheumatism (EULAR) criteria for diagnosis of RA [16, 17]. Participants were also required to have had a high or moderate disease activity at the start of bDMARD/tsDMARD treatment despite previous methotrexate (MTX) monotherapy for ≥3 months. Additionally, the patients were required to have already tapered off prednisolone at the start of bDMARD/tsDMARD discontinuation.

Discontinuation of bDMARDs and tsDMARDs

All participants started the bDMARD and tsDMARD discontinuation after maintaining remission or LDA for ≥48 weeks, and they were prospectively followed-up to assess disease activity. A direct discontinuation strategy without stepwise tapering was implemented. All patients were recommended to return for a follow-up visit every 8 weeks after bDMARD/tsDMARD discontinuation. They were allowed to receive a stable dose of MTX (6 to 12 mg/week) during follow-up. They were not allowed to use other conventional synthetic DMARDs (csDMARDs) or prednisolone to control disease activity during follow-up. Follow-up started on the day of bDMARD/tsDMARD discontinuation and ended at the earliest date of the following: flare occurrence, lost to follow-up, death, adverse events, or the last follow-up visit before July 1, 2021. Loss to follow-up was defined as missing at least two scheduled visits without any contact. Adverse events were defined as those that caused MTX discontinuation. Decisions to discontinue MTX due to adverse events were made by the treating physicians on the basis of a comprehensive evaluation of physical findings, laboratory findings, and radiological examinations at each visit or at any time during follow-up if patients had clinical signs or symptoms.

Rescue therapy

Patients who experienced a disease flare started rescue therapy, which comprised restarting the discontinued drugs (previous drugs). Disease activity was examined every 2 weeks during rescue therapy. When disease activity was not controlled by the previous drugs within 4 weeks, patients were treated with another bDMARD or tsDMARD. The choice of the other bDMARD or tsDMARD therapy was at the discretion of each treating physician.

DMARDs

The bDMARDs used in this study included the following two classes: tumor necrosis factor (TNF) inhibitors (infliximab, etanercept, adalimumab, golimumab, and certolizumab pegol) and an interleukin-6 inhibitor (tocilizumab). Tocilizumab was administered via an intravenous infusion at 8 mg/kg every 4 weeks or by a subcutaneous injection of 162 mg every other week. For patients who failed to achieve remission or LDA by the original subcutaneous tocilizumab regimen at 3 months, a 162-mg once-weekly regimen was used. In this study, tofacitinib was the only tsDMARD used to treat the participants. Tofacitinib was administered at a dose of 5 mg twice a day.

RA disease activity

The clinical disease activity index (CDAI) was used to quantify RA disease activity. Cut-off values for disease activity states were defined as follows: high disease activity, CDAI >22; moderate disease activity, CDAI >10 and ≤22; LDA, CDAI >2.8 and ≤10; and remission, CDAI ≤2.8 [18, 19]. After bDMARD/tsDMARD discontinuation, CDAI assessments were performed at each scheduled visit (usually every 8 weeks) or at any time if patients had clinical signs or symptoms. A disease flare was defined as a worsening of CDAI values (a return to moderate or high CDAI) at any time during follow-up.

Assessment of joint destruction

Radiographs of the hands and feet were taken at the start of treatment with bDMARD or tsDMARD, at the start of bDMARD/tsDMARD discontinuation, and every year during follow-up. These data were used to examine structural outcomes after bDMARD/tsDMARD discontinuation. Each radiograph was assessed independently by two readers (AO and TK) who were well trained and competent to score radiographs using the van der Heijde-modified total Sharp score (mTSS) [20]. The readers were also blinded to the patients’ clinical status and treatment and to the chronologic order of the films. To ensure objectivity of the scoring, the average of the mTSS scores determined by each of the two readers was used for each patient’s mTSS. A change in mTSS between the bDMARD/tsDMARD discontinuation and 3 years later (ΔmTSS) was determined for each patient who had maintained remission for 3 years or more. The proportion of patients with no (ΔmTSS ≤0 per 3 years) and little (ΔmTSS ≤1.0 per 3 years) radiological progression, which was defined as radiographic remission in this study, was determined [20].

Ethical approval

This study was conducted in accordance with the principles of the Declaration of Helsinki. The protocol of this study also met the requirements of the Ethical Guidelines for Medical and Health Research Involving Human Subjects, Japan and was approved by the Human Research Ethics Committees of National Hospital Organization Kumamoto Saishun Medical Center (Nos. 31–36 and 2–12). Informed written consent was obtained from all participants before enrollment into this study.

Statistical analysis

Mean and standard deviation (SD) were used as descriptive statistics for data with a continuous distribution, which included non-normally distributed data [21, 22], and as the number (percentage) for categorical data. Crude incidence rates (IRs) of flare and the 95% confidence intervals (CIs) were also calculated by dividing the number of incidence cases by the number of corresponding follow-up person-years (PYs).

The probability of a disease flare event occurring over time after drug discontinuation was estimated using the cumulative incidence function (CIF) because we considered the presence of competing risks in the present study (i.e., lost to follow-up). The occurrence of a competing risk event precludes the occurrence of the primary event of interest. In the presence of competing risks, the simple use of the Kaplan–Meier survival function can overestimate the cumulative incidence probability of a flare event. To avoid this possibility, we used the CIF instead of the Kaplan–Meier survival function. The median time to disease flare (95% CI) and the cumulative flare probability (95% CI) were estimated using the CIF. Gray’s test with the post hoc Holm’s procedure was used to test the equality of CIF plots among the previous drugs. Gray’s test is the analogue to the log-rank test that is used for testing the equality of Kaplan–Meier survival curves between groups [23, 24].

Considering the presence of competing risks, Fine‒Gray competing risk regression analysis was used to evaluate the effect of clinical characteristics on the occurrence of disease flare during follow-up after bDMARD/tsDMARD discontinuation. Clinical characteristics at the start of bDMARD/tsDMARD treatment and those at the start of drug discontinuation that were considered to be clinically relevant based on previous knowledge were used as predictor variables. Univariable Fine‒Gray regression analyses were performed first for each predictor variable. Thereafter, all predictor variables with p values <0.10 in the univariable analyses were introduced into a multivariable analysis. A backward selection procedure with a cut-off significance level of 0.05 was used in the multivariable model. Risk differences are presented as adjusted hazard ratios (HRs) with 95% CIs.

For all tests, a probability value (p value) of <0.05 was considered to indicate statistical significance. All calculations were performed using Easy R (Saitama Medical Center, Jichi Medical University, Saitama, Japan) [25].

Results

Patient characteristics

As shown in Fig 1, 97 patients with RA who desired drug discontinuation after maintenance of remission or LDA for ≥48 weeks due to stable treatment with bDMARDs (46 patients with TNF inhibitors and 19 with tocilizumab) or tofacitinib (32 patients) were enrolled into this study. The participants’ clinical characteristics are shown in Table 1. When treatment with these DMARDs was introduced, 44 patients (45.4%) had high CDAI values and 54 (55.7%) had early RA (≤2 years of disease duration). The mean mTSS was 16.9. The mean time from RA onset to the first use of MTX was 2.4 years, and 63 patients (64.9%) were bDMARD-naïve patients. During treatment with bDMARDs or tofacitinib, MTX was concomitantly used in 90 patients (92.8%), and 11 patients (11.3%) used prednisolone for priming when introducing bDMARDs or tofacitinib (within 3 months). Immediately before starting the bDMARD/tsDMARD discontinuation, 81 patients (83.5%) were in remission. The mean length of remission or LDA was 4.2 years. All patients received a stable dose of MTX after starting DMARD/tofacitinib discontinuation.

Fig 1

Disposition of patients: Discontinuation of bDMARDs and tofacitinib and rescue therapy.

*Patients who experienced a disease flare after discontinuation of TNF inhibitors (n = 32) achieved remission or LDA with the same TNF inhibitors (n = 24), abatacept (n = 1), or tofacitinib (n = 7). The other patients regained a good outcome with previous DMARDs. RA, rheumatoid arthritis; LDA, low disease activity; bDMARDs, biological disease-modifying antirheumatic drugs; TNFI, tumor necrosis factor inhibitor; TCZ, tocilizumab; ABT, abatacept; TOF, tofacitinib.

Table 1

Clinical characteristics of RA patients.

	Total (n = 97)
At the start of bDMARD or tofacitinib treatment
Age, years, mean (SD)	59.6 (11.5)
Male sex, number (%)	26 (26.8)
RA duration, years, mean (SD)	4.4 (7.6)
≤2 years, number (%)	54 (55.7)
Anti-CCP positive, number (%)	89 (91.8)
RF positive, number (%)	89 (91.8)
CDAI, mean (SD)	23.8 (11.3)
High (CDAI >22.0), number (%)	44 (45.4)
mTSS, mean (SD)	16.9 (33.2)
Erosion score, mean (SD)	7.8 (18.0)
No erosion, number (%)	27 (27.8)
Joint space narrowing score, mean (SD)	9.1 (15.9)
Time from onset to starting MTX use, years, mean (SD)	2.4 (7.1)
bDMARD naïve, number (%)	63 (64.9)
Treatment
TNF inhibitors, number (%)	46 (47.4)
Tocilizumab (IL-6 inhibitor), number (%)	19 (19.6)
Tofacitinib (Janus kinase inhibitor), number (%)	32 (33.0)
Concurrent use of MTX, number (%)	90 (92.8)
Initial use of PSL, number (%)	11 (11.3)
At drug discontinuation
Age, years, mean (SD)	63.9 (11.5)
RA duration, years, mean (SD)	8.5 (8.3)
CDAI, mean (SD)	1.96 (1.29)
Remission (CDAI ≤2.8), number (%)	81 (83.5)
Length of remission or low CDAI, years, mean (SD)	4.2 (2.9)

RA, rheumatoid arthritis; bDMARD, biological disease-modifying antirheumatic drug; TNF, tumor necrosis factor; IL-6, interleukin-6: MTX, methotrexate; PSL, prednisolone; anti-CCP, anti-cyclic citrullinated peptide antibodies; RF, rheumatoid factor; CDAI, clinical disease activity index; mTSS, van der Heijde-modified total Sharp score; SD, standard deviation

Disease flare following discontinuation of bDMARDs and tofacitinib

As shown in Table 2, the mean follow-up period after discontinuation was 2.1 years (SD 2.0), and disease flare occurred in 74 patients (76.3%). Crude IRs of disease flare during follow-up were 0.36 per PY (95% CI 0.29–0.45). CIF plots for the probability of disease flare from the discontinuation of bDMARDs and tofacitinib are shown in Fig 2A. The estimated median time to disease flare was 1.6 years (95% CI 0.9–2.6). The mean of the cumulative flare probability was estimated to be 0.45 (95% CI 0.35‒0.55) at 1 year, 0.56 (0.46‒0.66) at 2 years, 0.64 (95% CI 0.54‒0.73) at 3 years, and 0.80 (95% CI 0.68‒0.87) at 5 years (Table 2). Fig 2B shows the cumulative incidence of disease flare after grouping according to the discontinued drugs (TNF inhibitors, tocilizumab, and tofacitinib). There were significant differences in flare estimates over time among these patient groups (p = 0.038 for the comparison of all groups using Gray’s test and p = 0.023 for tofacitinib versus TNF inhibitors with the post hoc Holm’s test). The median time to disease flare was shorter in patients who had discontinued tofacitinib compared with TNF inhibitors (0.6 years vs. 2.2 years), and the flare rate at 1 year was greater in the tofacitinib group compared with the TNF inhibitor group (66% vs. 35%) (Table 2). Kaplan–Meier plots for the flare-free survival probability after the discontinuation of bDMARDs and tofacitinib are also shown in S1 Fig.

Table 2

Flare of disease activity in RA patients after drug discontinuation.

	Overall	TNF inhibitors	Tocilizumab	Tofacitinib
	(n = 97)	(n = 46)	(n = 19)	(n = 32)
Follow-up, years, mean (SD)*	2.1 (2.0)	2.4 (2.2)	2.4 (1.6)	1.6 (2.0)
Follow-up, years, median (IQR)*	1.2 (0.5‒3.6)	1.8 (0.8–4.0)	2.6 (0.7–3.7)	0.6 (0.3–2.2)
Flares, number (%)	74 (76.3)	32 (69.6)	15 (78.9)	27 (84.4)
Crude IR per PY (95% CI)	0.36 (0.29–0.45)	0.29 (0.20–0.41)	0.33 (0.20–0.56)	0.54 (0.37–0.79)
Estimates by CIF analysis
Time to disease flare years, median (95% CI) †	1.6 (0.9‒2.6)	2.2 (1.2–4.5)	3.1 (0.7–NA)	0.6 (0.3–1.6)
Probability of disease flare, mean (95% CI)
At 1 year	0.45 (0.35‒0.55)	0.35 (0.22–0.49)	0.42 (0.20–0.63)	0.66 (0.47–0.79)
At 2 years	0.56 (0.46‒0.66)	0.50 (0.34–0.64	0.42 (0.20–0.63)	0.78 (0.60–0.89)
At 3 years	0.64 (0.54‒0.73)	0.57 (0.41–0.71)	0.53 (0.29–0.72)	0.81 (0.63–0.91)
At 5 years	0.80 (0.68‒0.87)	0.77 (0.58–0.88)	–	0.81 (0.63–0.91)

*Follow-up was measured from the start of drug discontinuation.

†Median time to disease flare was the estimated time where 50% of patients would have a flare.

RA, rheumatoid arthritis; bDMARDs, biological disease-modifying antirheumatic drugs; IR, incidence rate; PY, person-year; CIF, cumulative incidence function; SD, standard deviation; IQR, interquartile range; 95% CI, 95% confidence interval

Fig 2

Cumulative incidence of disease flare after the discontinuation of bDMARDs and tofacitinib.

CIF plots for the probability of patients who experienced disease flare following drug discontinuation are shown for all patients (A) and patients grouped according to the previous drugs (B). The cumulative flare probability over time among the patient groups were compared using Gray’s test with the post hoc Holm’s procedure. The p values are as follows: p = 0.038 for a comparison among the three groups; p = 0.023 for tofacitinib versus TNF inhibitors; p = 0.14 for tocilizumab versus tofacitinib; and p = 0.53 for TNF inhibitors versus tocilizumab. Numbers below this figure represent the number of patients remaining in the analysis. CIF, cumulative incidence function; bDMARDs, biological disease-modifying antirheumatic drugs; TNF, tumor necrosis factor.

Rescue therapy for flares occurring after the discontinuation of bDMARDs and tofacitinib

As shown in Fig 1, all patients who experienced disease flare, except one who was lost to follow-up, started rescue therapy with the previous drugs. Among these patients, 89% recaptured remission or LDA as follows: all patients in the tocilizumab-discontinuation group (15 patients) and the tofacitinib-discontinuation group (26 patients) regained disease control within 1 month by restarting the previous drugs; among 32 patients who discontinued TNF inhibitors, 24 (75%) successfully achieved remission within 1 month of restarting the same drug. The other eight failed to achieve LDA, but remission was reached by switching to abatacept (one patient) or tofacitinib (seven patients).

Influence of bDMARD/tofacitinib discontinuation on structural outcome

Thirty-one patients maintained remission for 3 years or more after drug discontinuation (previous drugs: 16 TNF inhibitors, nine tocilizumab, and six tofacitinib). Mean values of mTSS, erosion score, and joint space narrowing score (JNS) were determined at the time of starting and discontinuing bDMARD/tofacitinib treatment and 3 years after drug discontinuation. A cumulative probability plot of 3-year changes in mTSS (ΔmTSS) from drug discontinuation is shown in Fig 3. No or little radiological progression was shown in 87.1% of the patients who maintained remission for 3 years or more after drug discontinuation; the rates of patients with ΔmTSS ≤0 per 3 years and those with ΔmTSS >0 and ≤1 per 3 years were 71.0% and 16.1%, respectively.

Fig 3

Structural outcome: Cumulative probability plot of the change in mTSS due to bDMARD/tofacitinib discontinuation over 3 years.

mTSS, van der Heijde-modified total Sharp score.

Predictive factors for the risk of disease flare following discontinuation of bDMARD and tofacitinib

Results of univariable and multivariable Fine‒Gray competing risk regression analyses are shown in Table 3. The univariable analyses showed that RA duration, mTSS score (erosion score), and time from RA onset to first MTX use at the start of bDMARD or tofacitinib treatment, previous failure of treatment with bDMARDs, the type of treatment drugs that led to remission or LDA, and disease activity and duration of maintaining the remission or LDA status at the start of bDMARD/tofacitinib discontinuation were predictor variables with p<0.10. These variables were placed into the multivariable regression analysis. In the multivariable regression model, longer RA duration, previous failure of treatment with bDMARDs, and a longer duration of remission or LDA status were identified as the predictive factors for disease flare; the adjusted HR (95% CI) was 1.06 (1.02‒1.10) for RA duration per additional year (p = 0.0041), 0.58 (0.35–0.94) for bDMARD-naïve compared with previous bDMARD failure (p = 0.028), and 0.89 (0.35–0.94) for duration of remission or LDA per additional year (p = 0.0083). After adjusting for other predictor variables, the treatment drug type did not remain a predictive factor for disease flare.

Table 3

Univariable and multivariable Fine‒Gray regression analyses for disease flare after drug discontinuation.

Variables	Unadjusted HR (95% CI)	p	Adjusted HR (95% CI)	p
At the start of bDMARD or tofacitinib treatment
Age per additional year	1.00 (0.98–1.02)	0.98	–	–
Male vs. female	0.99 (0.58–1.70)	0.97	–	–
RA duration per additional year	1.05 (1.01–1.09)	0.026	1.06 (1.02–1.10)	0.0041
Anti-CCP positive vs. negative	1.87 (0.57–6.10)	0.30	–	–
RF positive vs. negative	1.95 (0.62–6.11)	0.25	–	–
CDAI per additional unit	1.01 (0.99–1.02)	0.35	–	–
mTSS per additional unit	1.00 (1.00–1.01)	0.083	–	–
Erosion score per additional unit	1.01 (1.00–1.01)	0.051	–	–
No erosion vs. presence of erosion	1.14 (0.67–1.92)	0.63	–	–
Joint space narrowing score per additional unit	1.01 (1.00–1.02)	0.19	–	–
Time from onset to starting MTX use per additional year	1.05 (1.00–1.10)	0.051	–	–
bDMARD naïve vs. previous bDMARD failure	0.61 (0.37–0.98)	0.043	0.58 (0.35–0.94)	0.028
Drug treatment
Drugs
TNF inhibitors (reference)	1 (reference)	–	–	–
Tocilizumab	1.15 (0.67–1.97)	0.62	–	–
Tofacitinib	1.83 (1.05–3.20)	0.034	–	–
MTX use vs. no use	1.01 (0.35–2.91)	0.99	–	–
Initial PSL use vs. no use	0.97 (0.42–2.22)	0.93	–	–
At drug discontinuation
Age per additional year	1.00 (0.98–1.02)	0.72	–	–
RA duration per additional year	1.02 (0.98–1.07)	0.30	–	–
CDAI per additional unit	1.02 (0.86–1.21)	0.84	–	–
Remission vs. low CDAI	0.59 (0.31–1.10)	0.096	–	–
Length of remission or low CDAI per additional year	0.93 (0.86–1.01)	0.076	0.89 (0.82–0.97)	0.0083

All predictor variables with p values lower than 0.10 in the univariable Fine‒Gray model were introduced into multivariable analysis. For mTSS, erosion score was used. A backward stepwise selection procedure with a cut-off significant level of 0.05 was used in the multivariable model. Adjusted HRs (95% CIs) are shown for variables that remained in the final Fine‒Gray model.

RA, rheumatoid arthritis; bDMARD, biological disease modifying antirheumatic drug; TNF, tumor necrosis factor; MTX, methotrexate; PSL, prednisolone; anti-CCP, anti-cyclic citrullinated peptide antibodies; RF, rheumatoid factor; CDAI, clinical disease activity index; mTSS, van der Heijde-modified total Sharp score; HR, hazard ratio; 95% CI, 95% confidence interval

Safety

There were no severe adverse events that caused MTX withdrawal during the follow-up after bDMARD/tofacitinib discontinuation.

Discussion

Following bDMARD and tofacitinib discontinuation after the sustained remission or LDA status for ≥48 weeks in RA patients, disease flare occurred at an IR of 0.36 per PY, and the median time to flare was 1.6 years. No or little radiological progression was observed in 87.1% of patients who maintained remission for 3 years. In patients with a flare, 89% regained remission or LDA within 1 month by restarting the previous treatment regimens. The Fine‒Gray competing risk regression analysis showed that a longer RA duration at the start of bDMARD/tofacitinib, previous failure of treatment with bDMARDs, and shorter sustained duration of remission or LDA were predictive factors for the occurrence of a disease flare after bDMARD/tofacitinib discontinuation.

In daily clinical practice, patients frequently desire drug discontinuation after reaching remission or LDA status due to the treatment with bDMARDs or tsDMARDs because of the fear of adverse events associated with these drugs. When considering drug discontinuation, one of the most critical concerns is whether patients can rapidly recover their baseline condition by reintroducing the previous DMARDs if a disease flare occurs. In the present study, 89% of patients who experienced disease flare regained disease control using rescue therapy with the previous DMARDs. Successful recapture of remission or LDA by restarting the previous regimen has been reported in recent RCTs and observational studies involving bDMARDs and tsDMARDs [13–15, 26–37]. Rapid recovery of disease control with rescue therapy may encourage rheumatologists and RA patients to make a therapeutic decision in favor of bDMARD/tsDMARD discontinuation. Patients need to be informed about the risk of disease flares and their management strategies [38]. Additionally, patient eligibility for discontinuation should be considered. In this study, patients were scheduled to start bDMARD and tsDMARD discontinuation if they maintained a CDAI-based remission or LDA for at least 48 weeks with stable bDMARD or tsDMARD treatment and had already tapered off prednisolone.

Henaux et al. performed a systematic review and meta-analysis on structural damage, and they showed that bDMARD discontinuation was more likely to increase the risk of losing remission or LDA as well as the risk of radiological progression compared with continuing the initial bDMARD regimen [12]. In the PRESERVE study, patients who started discontinuation of etanercept had disease flares and showed radiological progression at 52 weeks more frequently compared with those receiving full- or reduced-dose etanercept [39]. In the DRESS study, radiological progression was associated with a time-weighted average of disease activity for 18 months after tapering of TNF inhibitors, although there was no association between the occurrence or number of flares and radiological progression. The findings suggested that radiological progression is caused by the long-term increase in disease activity over time after tapering [40]. These studies indicated the importance of maintaining a state of remission or LDA in patients as well as the need to regularly monitor radiological progression. It is not clear whether small increases in joint damage may become significant over time. In the present study, we showed that radiological damage did not continue to progress in approximately 90% of RA patients who maintained remission over 3 years after bDMARD/tofacitinib discontinuation.

There are few long-term outcome data regarding the bDMARD/tsDMARD-free strategy. In the HONOR study, which was conducted in Japan, 52 patients who maintained remission with adalimumab plus MTX for at least 6 months started adalimumab discontinuation. During a 5-year follow-up, adalimumab-free remission, which was determined by the 28-joint disease activity score based on erythrocyte sedimentation rate, was persistent in 48% of patients at 1 year, 31% at 3 years, and 21% at 5 years. No significant radiological changes were observed in patients who sustained LDA for 5 years, and among them, eight of nine patients (88.9%) maintained structural remission defined as ΔmTSS ≤0.5 per year [37]. In the present study, the rates of flare patients were estimated to be 45% at 1 year, 64% at 3 years, and 80% at 5 years after the discontinuation of bDMARDs and tofacitinib in RA patients. Among patients achieving bDMARD/tofacitinib-free 3-year remission, 27 of 31 (87.1%) had an mTSS ≤1 over the 3 years. Similar findings were obtained between the HONOR study and the present study.

In the present study, the predictive factors for disease flare following bDMARD/tofacitinib discontinuation were longer RA duration at the start of bDMARD/tofacitinib treatment, previous failure of biological treatment, and a shorter sustained period of the remission or LDA status before the start of drug discontinuation. Previous studies also identified several patient clinical and imaging characteristics as determinant factors for failure or success to sustain remission or LDA after the start of bDMARD tapering and discontinuation. The identified characteristics included disease activity at baseline, RA duration, timing of treatment (initial or delayed), previous use of bDMARDs, residual synovial inflammation under ultrasound examinations, and deep/long-standing remission [26, 37, 41–48]. These patient characteristics may help predict the sustained clinical benefit after bDMARD/tofacitinib tapering and discontinuation.

There are several limitations to this study. First, the sample size was relatively small, which may have made it difficult to draw a definitive conclusion about predictors for disease flare after bDMARD/tofacitinib discontinuation. To estimate the sample size, we needed to input an expected probability of flare-free survival at a time point of interest. However, we were unable to determine the expected long-term survival probability due to a lack of previous data. Second, the present study was an observational study in a real-world clinical setting, and therefore, allowing bDMARD/tofacitinib discontinuation was left to the discretion of the treating physician and the patient’s preference, which may have produced some bias. However, we used clear inclusion criteria for this study. We included only patients with MTX-resistant RA who had sustained remission or LDA with stable DMARD treatment (which included the type and dose) for ≥48 weeks and had already tapered off prednisolone. Additionally, we defined a disease flare based on CDAI values that were regularly monitored at each scheduled visit. We continued stable doses of MTX for all patients, but an additional use of prednisolone or other csDMARDs was not allowed to control disease activity during follow-ups. Considering such consistency in the definition of a flare and the eligibility of participants, the present study provides useful information regarding long-term outcomes in RA patients after bDMARD/tofacitinib discontinuation.

Conclusion

In this longitudinal observational study, it was estimated that half of RA patients had disease flare within 1.6 years after discontinuing bDMARDs and tofacitinib. The timing of bDMARD/tofacitinib start (RA duration before the treatment), previous failure of treatment with bDMARDs, and the length of remission or LDA before the discontinuation were significant predictive factors for successful discontinuation. Approximately 90% of patients who maintained bDMARD/tofacitinib-free remission for 3 years had no or little radiological progression. Restarting the original treatment regimen allowed rapid recapture of disease control within 1 month in almost all flare cases. Thus, attempting bDMARD/tofacitinib-free management may be a viable option that could be pursued after achieving stable control of disease activity in RA patients, especially those who had started the treatment at early RA stages and maintained remission or LDA for a longer time. Regular monitoring of disease activity and patient education about early signs of disease flare are crucial to prevent persistent disease flares and radiological damages during bDMARD/tofacitinib-free treatment for RA.

Kaplan–Meier plots: The probability of flare-free survival after the discontinuation of bDMARDs and tofacitinib.

Kaplan–Meier plots for the proportion of patients who experienced no disease flare during follow-up are shown for patients who discontinued TNF inhibitors, tocilizumab, or tofacitinib. The flare-free probability among the patient groups was compared using the log-rank test with the post hoc Holm’s procedure. The p values were as follows: p = 0.069 for a comparison among the three groups; p = 0.030 for tofacitinib versus TNF inhibitors; p = 0.14 for tocilizumab versus tofacitinib; and p = 0.52 for TNF inhibitors versus tocilizumab. Median time to disease flare (95% CI) was 2.2 years (1.0–3.4 years) for TNF inhibitors, 3.1 years (0.0–6.3 years) for tocilizumab, and 0.6 years (0.2–1.0 years) for tofacitinib. Numbers below this figure represent the number of patients remaining in the analysis. bDMARDs, biological disease-modifying antirheumatic drugs; TNF, tumor necrosis factor; CI, confidence interval.

(TIF)

Click here for additional data file.

4 Apr 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Chengappa Kavadichanda

Academic Editor

PLOS ONE

Journal Requirements:

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Major issues:

Please clarify if this was a prospective cohort.

There is little data on how dDMARDs or ts DMARDs were discontinued. Where these tapered off or stopped suddenly? Was any other csDMARD allowed except MTX?

What about loss to follow-up? How was it ensured that the patients would be coming back every month (for CDAI) for such long periods?

Abstract:

It should mention the median (IQR) time on follow-up.

Short title: Should be mention relapse after bDMARD/tofacitinib discontinuation?

Introduction:

“ Although there is currently no cure…” sounds odds. What is cure if not sustained drug-free remission?

Methods:

Radiographs were obtained at what frequency?

How was normality of data checked?

Results:

Please provide SD for mean follow-up period.

How was “severe adverse effects” screened for?

Could a Kaplan Meier curve be added with log rank tests between Tofacitinib and bDMARDs?

Reviewer #2: Summary In this observational longitudinal cohort study. The investigators have looked at the impact of discontinuation of biological DMARDs or tofacitinib on Rheumatoid arthritis patients who were in clinical remission or having low disease activity.

• As if now the pointers towards drug tapering in patients with RA are not clear but studies suggest its feasibility

• This study focuses on the variables associated with increased risk of disease flare after treatment discontinuation

• The study is highly appreciable and have been done in a more meaningful way

First Impression

The research is original as well as relevant in the clinical practice and may help practitioners in stratifying their patients with RA with relatively higher risk of relapse after treatment discontinuation and to execute a better screening for considering drug tapering or stopping once they achieve remission.

• The article has been structured well and articulated in a lucid manner.

Abstract • Explicitly written, well summarized, involves the salient findings of the study.

Following point –

1. It will be nice if median duration of follow up can be mentioned in abstract

Introduction

• The introduction is well structured and provide the correct insight to the purpose of the study as well the assumed hypothesis.

Methodology

• The methodology and the work flow described is clear and easy to follow.

Following points –

1. Line 98 page 12“First-date” could be possibly rewritten as “since the day of drug discontinuation”

2. Line 103 page 12 - “When disease activity was not controlled by the previous drugs, patients were treated with

another bDMARD or tsDMARD” what was the waiting period before starting a new drug

3. Line 112 page 13 – “For patients who were refractory to” how was this refractory state defined.

4. What was the degree of agreement between two blinded readers assessing radiographs?

Statistics

• Lucid description and appropriate methods mentioned for result analysis.

Following points-

1. How sample size was calculated for this study

Results and discussion

• Organized and well-structured.

Following points –

1. Line 173 page 15 onwards – kindly also mention absolute numbers of the patients (only percentage has been mentioned)

2. Line 184 to 186 page 15 – Mention “Table 02” somewhere since these findings are summarized in same

Conclusion

• Structured to the point and gives the summary of the primary finding.

Figures and tables

• All the figures and tables are clear and self-explanatory.

Following points –

1. Few variables mentioned as mean and standard deviation appears to have standard deviation more than mean, probably because of skewed data. Such variables may be more appropriately mentioned as median with Interquartile range or range rather than mean and standard deviation

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

28 Apr 2022

Response to Reviewers

We are grateful to the reviewers for their valuable comments as well as their time and energy that they put into reviewing our manuscript. We have made all requested changes and added new information to the manuscript in response to their insightful comments. All alterations are highlighted in red text in the revised version of the manuscript. We are confident that the manuscript has benefited from the reviewers’ useful comments and suggestions.

Below are point-by-point replies to the reviewers’ comments.

Reply to Reviewer 1

Major issues

1. We wish to thank the reviewer for the comment regarding the study design. In this study, all participants who underwent bDMARD/tofacitinib discontinuation were prospectively followed-up until disease flare or until the end of the study. To clarify this point, we included the word “prospective” in the title and revised the title as follows: “Long-term outcomes after discontinuing biological drugs and tofacitinib in patients with rheumatoid arthritis: a prospective cohort study” (lines 2 and 3). We also added the word “prospectively” to several sentences in the abstract (line 33) and the Patients and Methods section (line 96).

2. We appreciate the reviewer’s comment that the strategy for bDMARDs/tofacitinib discontinuation should be clarified. In the present study, we used a direct DMARD discontinuation strategy instead of a gradual discontinuation strategy for patients who desired drug discontinuation after maintaining remission or low disease activity (LDA) for ≥48 weeks through stable drug treatment with bDMARD or tofacitinib. Although recent clinical studies have suggested that stepwise tapering until complete discontinuation of DMARDs could be safely implemented with little risk of flares, the optimal strategy for gradual withdrawal for each DMARD remains unknown. To clarify this point, several sentences were added to the revised manuscript (lines 95–98). As mentioned in the Results and Discussion sections of the previous version of our manuscript, we continued a stable dose of methotrexate (MTX) for all patients during follow-up, but additional use of prednisolone or other csDMARDs was not allowed to control disease activity (lines 210, 211, and 356–358). To clarify this point, we added this information to the Patients and Methods section of the revised manuscript (lines 100–102).

3. We apologize for the error. After bDMARD/tofacitinib discontinuation, clinical disease activity was not measured every month, but, rather, at each scheduled visit (usually every 8 weeks) or at any time during follow-up if patients had clinical signs or symptoms. In our rheumatology department, all patients need to make an appointment for next medical examinations during follow-up. We recommended that the patients should return for a follow-up visit every 8 weeks after bDMARD/tofacitinib discontinuation. If patients missed at least two scheduled visits without any contact, they were considered to be lost to follow-up. For patients who experienced a disease flare after bDMARD/tofacitinib discontinuation, disease activity was examined every 2 weeks during rescue therapy. We included this information in the revised manuscript (lines 98, 99, 104, 105, 114, 115, 134–136, 355, and 356).

Abstract

1. We wish to thank the reviewer for the comment that follow-up duration should be described in the abstract. The median (interquartile range [IQR]) length of follow-up was 1.2 years (0.5–3.6 years). However, recent recommendations for statistical analysis in rheumatology studies state that the mean and standard deviation (SD) are meaningful descriptive statistics for data following continuous distributions, even for non-normally distributed data (Recommendation #9: Lydersen S. Statistical review: frequently given comments. Ann Rheum Dis 2015; 74, 325-325; Lydersen S. Mean and standard deviation or median and quartiles? Tidsskr Nor Laegeforen 2020; 140(9), pii 20-0032). Additionally, an in-depth discussion on the “mean and SD” was published by Professor Lydersen in 2021 (Ann Rheum Dis 2021; doi:10.1136/annrheumdis-2021-220336). In the present study, in accordance with this recommendation, we used the mean and SD as descriptive statistics for all data with continuous distributions except the median time to disease flare, which was computed in the survival analysis. We added a mean follow-up of 2.1 years (SD 2.0) to the abstract (lines 35 and 36 in the abstract). This information on descriptive statistics was added to the Patients and Methods section (lines 164–166). New references were provided (refs. 21 and 22). We understand that the median (IQR) is often used to show the distribution of follow-up time data, and we also included these data in Table 2.

Short title

1. We appreciate the reviewer’s suggestion to use in the short title “Relapse after bDMARD/tofacitinib discontinuation.” In the present study, we indicated that, besides the risk of disease flare after bDMARD/tofacitinib discontinuation, restarting the previous treatment regimens led to rapidly regaining disease control in 89% of flare patients within 1 month. Flares are manageable in most patients with rheumatoid arthritis. Considering this point and the reviewer’s suggestion, we revised the short title to “Disease flare and rescue therapy after bDMARD/tofacitinib discontinuation” (line 25).

Introduction

1. We understand the reviewer’s concern regarding the description “although there is currently no cure.” This sentence may lead to a misunderstanding by the reader. We deleted this sentence in the revised manuscript.

Patients and Methods

1. We appreciate the reviewer’s comment regarding the assessment of joint destruction. In this study, we took radiographs of the hands and feet at the start of bDMARD or tofacitinib treatment, at the start of bDMARD/tofacitinib discontinuation, and every year during follow-up after discontinuation. A change in van der Heijde-modified total Sharp score (mTSS) between the bDMARD/tsDMARD discontinuation and 3 years later (ΔmTSS) was determined for each patient who had maintained remission for 3 years or more. The average of the mTSS scores that were determined by two readers was used for each patient’s mTSS. We added this information to the revised manuscript (lines 140–142 and 147–150).

2. We appreciate the reviewer’s comment on the normality of the mTSS data. As the reviewer pointed out, the data did not follow a normal distribution. As we mentioned in the response to Comment #1 about the abstract, however, recent recommendations for statistical analysis in rheumatology studies state that, as descriptive statistics, the mean (SD) is meaningful and well defined for data with continuous distributions, even for non-normally distributed data (refs. 21 and 22). Additionally, in recent studies, some of which were cited in this manuscript, the mean and SD were used for the mTSS/ΔmTSS data (e.g., refs 29, 33, 37, 39, and 40). Thus, we used the mean (SD) for mTSS and ΔmTSS (Table 1 and Fig. 3).

Results

1. In accordance with the reviewer’s comment, we added SD for the mean follow-up period (lines 214 and 215). The data are also shown in Table 2.

2. We appreciate the reviewer’s comment regarding screening for severe adverse events. As mentioned in the Patients and Methods section of the previous manuscript, all participants were allowed to receive a stable dose of MTX after discontinuing bDMARDs and tofacitinib during follow-up (lines 99 and 100). Severe adverse events were defined as those that caused MTX discontinuation during follow-up. Decisions to discontinue MTX due to adverse events were made by the treating physicians based on a comprehensive evaluation of physical findings, laboratory findings, and radiological examinations at every visit or at any time when patients had clinical signs or symptoms. We included this information in the Patients and Methods section of the revised manuscript (lines 106–110 and 273).

3. We appreciate the reviewer’s suggestion to add Kaplan–Meier (K–M) plots with the log-rank test. In the present study, we used the cumulative incidence function (CIF) to estimate the probability of the occurrence of a flare event over time, because we considered the presence of competing risks. As the reviewer pointed out in Major Issues Comment #3, there were lost to follow-up events in this study, which were treated as competing risk events in survival analysis. The occurrence of a competing risk event precludes the occurrence of the primary event of interest. In the presence of competing risks, the simple use of the K–M survival function can overestimate the cumulative incidence probability of a flare event. To avoid this possibility, we used the CIF instead of the K–M survival function. Gray’s test for the CIF model was used to test the equality of CIF plots among patient groups. Gray’s test is the analogue to the log-rank test that is used to test the equality of K–M survival curves between groups. To clarify this point, we added this information to the Patients and Methods section of the revised manuscripts (lines 169–180). New references were also provided (refs. 23 and 24). We agree with the reviewer’s suggestion that it may be useful to provide both CIF plots and K–M plots. Thus, we added K–M plots for the flare-free survival probability after bDMARD/tofacitinib discontinuation in the revised manuscript (lines 228–230, supporting information lines 378–392 and Fig S1).

Reply to Reviewer 2

We appreciate the reviewer’s comments about our study as a whole.

Abstract

1. We thank the reviewer for the suggestion to include the median duration of follow-up in the abstract. The median duration of follow-up (interquartile range [IQR]) was 1.2 years (0.5–3.6 years). However, recent recommendations for statistical analysis in rheumatology studies state that the mean and standard deviation (SD) are meaningful and well defined for data following all types of continuous distributions, even for non-normally distributed data (Lydersen S. Statistical review: frequently given comments. Ann Rheum Dis 2015; 74, 325-325; Lydersen S. Mean and standard deviation or median and quartiles? Tidsskr Nor Laegeforen 2020; 140, pii: 20-0032). We therefore included the mean follow-up (SD) in the abstract of the revised manuscript (lines 35 and 36). We also added an explanation that the mean and SD were used as descriptive statistics for data with continuous distributions, even for non-normally distributed data in the Patients and Methods section (lines 164–166). New references were provided (refs. 21 and 22). We understand that median (IQR) is often used to show the distribution of follow-up time data, and we also included these data in Table 2.

Methodology

1. In accordance with the reviewer’s comment, we revised this sentence to “Follow-up started on the day of bDMARD/tsDMARD discontinuation” (lines 102 and 103).

2. We appreciate the reviewer’s comment regarding the waiting period before the start of a new DMARD. Patients who experienced a disease flare started rescue therapy with previous DMARDs. Disease activity was examined every 2 weeks during rescue therapy. We changed previous bDMARDs or tsDMARDs to another drug if disease activity was not controlled by the previous drugs by 4 weeks. We added this information to the Patients and Methods section (lines 114–116).

3. We appreciate the review’s comment on the definition of “refractory to the original subcutaneous regimen of tocilizumab.” For patients who failed to achieve remission or low disease activity (LDA) by the original subcutaneous regimen at 3 months, a 162-mg once-weekly regimen was used. We added this information to the revised manuscript (lines 125–127).

4. We thank the review for the comment regarding the degree of agreement between two blinded readers who scored the radiographs of the hands and feet. These two readers (Drs. Okada and Koga) completed a training course on the Sharp-van der Heijde method at Leiden University Medical Center and received a Certificate of Completion that was issued by Professor van der Heijde. In this study, they independently scored each radiograph using the van der Heijde-modified total Sharp score (mTSS). To ensure objectivity of the scoring, the average of the mTSS scores determined by each of the two readers was used for each patient’s mTSS. We included this information in the Patients and Methods section of the revised manuscript (lines 147 and 148).

Statistics

1. We appreciate the reviewer’s comment on sample size estimation. To estimate the sample size in a survival study for time-to-event data, we needed to input the expected probability of event-free survival at a time point of interest. The present study was a prospective follow-up to evaluate long-term outcomes of bDMARD/tofacitinib discontinuation. We estimated an appropriate probability of disease flare/probability of flare-free survival after direct bDMARD/tofacitinib discontinuation based on previous data on the long-term survival probability. As mentioned in the Introduction section, however, follow-up periods after discontinuation of drugs were within 1 year in most studies (lines 73 and 74). There are few long-term outcome data regarding the bDMARD/tsDMARD-free strategy after remission or LDA achievement. In the HONOR study, which was an open-label non-randomized study, 52 patients who maintained remission with adalimumab plus MTX for at least 6 months started adalimumab discontinuation. During a 5-year follow-up, adalimumab-free remission was persistent in 21% of patients at 5 years (a flare-free survival probability of 0.21 at 5 years) (ref. 37, lines 319–325). However, the inclusion criteria (the status of disease activity and its duration before starting adalimumab discontinuation) were different from ours. Additionally, long-term survival data regarding other tumor necrosis factor inhibitors, tocilizumab, or tofacitinib (e.g., flare-free survival probability at 5 years after drug discontinuation) were not available in the literature. It was difficult to estimate a long-term probability of disease flare or flare-free survival due to a lack of previous data. We were therefore unable to estimate the sample size for this study. We hope our long-term survival data will be helpful to calculate sample size for future studies. We added an explanation that we could not estimate the sample size for this study due to the lack of previous data in the Discussion section of the revised manuscript (lines 346–349).

Results and Discussion

1. We agree with the reviewer’s suggestion to include the absolute patient number in this section. Thus, we changed “percentage” to “patient number (percentage)” in the revised version of the text. These data are also shown in Table 1 (lines 201–209).

2. In accordance with the reviewer’s comment, we included “As shown in Table 2” at the beginning of this sentence in the revised manuscript (line 214).

Figures and Tables

1. We understand the reviewer’s concern regarding the use of the mean and SD for a skewed data set. As we mentioned in the response to Comment #1 about the abstract, however, recent recommendations for statistical analysis in rheumatology studies state that as descriptive statistics, the mean and SD are well defined and relevant for data with continuous distributions, even for non-normally distributed data (refs. 21 and 22). Additionally, an in-depth discussion on the “mean and SD” was published by Professor Lydersen in 2021 (Ann Rheum Dis 2021; doi:10.1136/annrheumdis-2021-220336). Recent rheumatology studies, some of which are presented in this study, used the mean and SD for baseline patient characteristics even though they were skewed distributed data. Considering these points, we used the mean and SD as descriptive statistics for all data with continuous distributions except the median time to disease flare, which was computed in the survival analysis. As mentioned above, we added this information to the Patients and Methods section (lines 164–166).

Submitted filename: Response to Reviewers.doc

Click here for additional data file.

10 Jun 2022

Long-term outcomes after discontinuing biological drugs and tofacitinib in patients with rheumatoid arthritis: a prospective cohort study

PONE-D-22-04611R1

Dear Dr. Mori,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Chengappa Kavadichanda

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for the clarifications and corrections.

Thank you for the clarifications and corrections.

Thank you for the clarifications and corrections.

Thank you for the clarifications and corrections.

Thank you for the clarifications and corrections.

Thank you for the clarifications and corrections.

Reviewer #2: Thank you for the adequate response to all the comments. All comments have been addressed adequately

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

13 Jun 2022

PONE-D-22-04611R1

Long-term outcomes after discontinuing biological drugs and tofacitinib in patients with rheumatoid arthritis: a prospective cohort study

Dear Dr. Mori:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Chengappa Kavadichanda

Academic Editor

PLOS ONE