Abstract
Objective This study examined the effect of dose step-down in patients with rheumatoid arthritis (RA) who achieved sustained disease control with baricitinib 4 mg once daily for up to 96 weeks.
Methods Patients who completed a baricitinib phase III study could enter a long-term extension (LTE). In the LTE, patients who received baricitinib 4 mg for ≥ 15 months and maintained Clinical Disease Activity Index (CDAI) low disease activity (LDA) or remission (REM) were blindly randomized to continue with 4 mg or taper to 2 mg. If needed, 2-mg–treated patients could be rescued to 4 mg, and 4-mg–treated patients could be rescued by adding or increasing conventional synthetic disease-modifying antirheumatic drugs. Efficacy and safety were assessed through 96 weeks. Nonresponder imputation (NRI), considering rescued or discontinued patients as nonresponders, was used for CDAI response analyses.
Results At 96 weeks, most patients maintained LDA in both 2-mg and 4-mg arms, with a lower maintenance rate in the 2-mg than the 4-mg group (NRI 59.9% and 70.2%, respectively). Patients maintained REM in the 2-mg and 4-mg arms at 30.8% and 36.6%, respectively. Rescue rates were 14.7% for baricitinib 4 mg and 22.5% for 2 mg. Of 112 patients who lost LDA in the 2-mg arm and were rescued to 4 mg, 76.2% and 75.6% achieved LDA again at 12 and 24 weeks postrescue, respectively.
Conclusion In a randomized, blinded, phase III LTE study, maintenance of disease control following induction of sustained LDA/REM with baricitinib 4 mg was greater with continued 4 mg than after tapering to 2 mg. Nonetheless, 76% of patients tapered to 2 mg could maintain LDA/REM or recapture with return to 4 mg, if needed.
Rheumatoid arthritis (RA) is a chronic inflammatory joint disease that can cause cartilage and bone damage as well as disability.1 Clinical recommendations suggest that the treatment target for RA should be sustained clinical remission (REM).2-4 International guidelines recommend the tapering of disease-modifying antirheumatic drugs (DMARDs) in patients with RA with sustained REM or satisfactory disease control.2-4 DMARD tapering is an area of emerging interest in clinical studies5,6; however, evidence on this treatment strategy remains limited for Janus kinase (JAK) inhibitors.
Baricitinib is an oral selective and reversible inhibitor of JAK1 and JAK2, which modulate signal transduction for a variety of cytokines involved in the immune-inflammatory response.7 Baricitinib was approved for the treatment of moderately to severely active RA in adults who have responded inadequately or who are intolerant to ≥ 1 DMARDs in > 75 countries, including European nations, the USA, and Japan.8-11
In a long-term extension study of baricitinib (RA-BEYOND; ClinicalTrials.gov: NCT01885078) in patients with RA who completed phase III studies (prior studies), patients who achieved sustained low disease activity (LDA) receiving baricitinib 4 mg were rerandomized to baricitinib 4 mg (continued) or 2 mg (step-down) in a double-blinded manner.5 Published preliminary data reported approximately 80% of patients in the baricitinib 4-mg group and 70% of patients in the 2-mg group sustained LDA for 48 weeks. In patients receiving 4 mg once daily who achieved sustained LDA, some were able to reduce baricitinib to 2 mg once daily.5 The study also found that most patients maintained control of disease activity, or regained it with return to 4 mg if needed.
Preliminary data were limited because of small sample size and short duration of dose tapering up to 48 weeks. The completed RA-BEYOND study therefore provided an opportunity for our study to evaluate the effects of baricitinib dose reduction in target patients who achieved sustained disease control with baricitinib 4 mg with a longer period of dose tapering (up to 96 weeks) and an increased sample size.
METHODS
Study design.
• RA-BEYOND. Study participants were originally patients enrolled in 4 phase III studies and had an inadequate response or intolerance to DMARDs (ClinicalTrials.gov: NCT01721057 [RA-BUILD]; NCT01721044 [RA-BEACON]; NCT01710358 [RA-BEAM]; and NCT04449224 [RA-BALANCE]). Patients who completed these studies were eligible to enter RA-BEYOND (NCT01885078) to evaluate the long-term safety and efficacy of baricitinib. In the originating studies, patients who were nonresponsive at week 16 were rescued. If they continued to be nonresponsive, they were discontinued from the studies. Laboratory abnormalities or significant uncontrolled medical conditions deemed to be a risk to baricitinib administration precluded participation in RA-BEYOND.
In RA-BEYOND, patients receiving 4-mg and 2-mg doses continued to receive the concluding dose of their originating study. Patients were switched to baricitinib 4 mg upon entry if they had received a placebo or active comparator at the end of their original study. Treatment assignment was not disclosed to patients and investigators. Patients were permitted to continue any background noninvestigational, open-label conventional synthetic DMARDs (csDMARDs), nonsteroidal antiinflammatory drugs, or corticosteroids they were receiving. Rescue therapy was allowed for any patient who had a Clinical Disease Activity Index (CDAI) score > 10 at or after 3 months following enrollment in RA-BEYOND. Patients receiving baricitinib 2 mg would be rescued by increasing the dose to open-label 4 mg and those already on 4 mg would be rescued by adding or increasing methotrexate or other csDMARDs. CDAI evaluation does not require inflammatory markers, and therefore permits immediate determination of the need of rescue or eligibility for the randomization substudy.
• Randomization substudy. Patients enrolled in RA-BEYOND were eligible to participate in the step-down randomization substudy if they received baricitinib 4 mg for ≥ 15 months (including time in the originating study) and achieved sustained LDA (defined by CDAI score ≤ 10 at 2 consecutive visits ≥ 3 months apart). Prior rescue in the original study or RA-BEYOND excluded patients from step-down eligibility. The eligibility criteria are illustrated in Supplementary Figure S1 (available with the online version of this article).
Patients meeting eligibility for substudy participation were rerandomized 1:1, stratified by geographic region and originating study, to maintain baricitinib 4 mg or to step down to 2 mg. The randomization occurred through an interactive web-based system without knowledge of the investigators or the patients. Patients receiving baricitinib 2 mg received 4 mg placebo and patients receiving baricitinib 4 mg received 2 mg placebo. Within the step-down substudy, investigators could provide rescue at any time for patients who failed to retain LDA.
Efficacy evaluation for step-down substudy. Two primary endpoints were examined. The first was the proportion of patients who maintained a CDAI score of ≤ 10 in the DMARD–inadequate responder (IR) population at weeks 12, 24, 48, and 96 of treatment with baricitinib 2 mg daily compared with patients continuing treatment with 4 mg daily. LDA was defined as CDAI ≤ 10 and REM as CDAI ≤ 2.8. The second primary endpoint was time to relapse (defined as a CDAI score > 10) after randomization to baricitinib 2 mg or continuation on 4 mg in this population. Additional analyses included assessments of change from the time of step-down randomization in composite scores (eg, Simplified Disease Activity Index [SDAI], Disease Activity Score in 28 joints [DAS28]–high-sensitivity C-reactive protein [hsCRP], DAS28–erythrocyte sedimentation rate [ESR], Health Assessment Questionnaire–Disability Index [HAQ-DI]) and their components.
Safety evaluation in step-down substudy. Treatment-emergent adverse events (TEAEs), serious adverse events (AEs), and AEs leading to permanent discontinuation of study drug were summarized in the 2-mg and 4-mg arms.
Statistical analyses. The substudy’s modified intention-to-treat population included all patients randomized into the step-down substudy ≥ 96 weeks prior to the data cut-off date (Supplementary Figure S2, available with the online version of this article) and had received ≥ 1 dose of the study drug after randomization. There was no prospective assessment of sample size and statistical power for efficacy analysis in the substudy. However, a sample size of 498 patients in each treatment group gives > 88% of power to detect a true difference of 10% between groups using a 2-sided Fisher exact test at a significance level of 0.05.
Fisher exact tests assessed treatment comparisons of CDAI response rates. The Kaplan-Meier method examined difference in time to event (ie, relapse and rescue) between treatment groups. Treatment comparisons of continuous efficacy endpoints were examined using t tests. No multiplicity control was applied for endpoints assessed in this substudy. All statistical tests were performed at a 2-sided significance level of 0.05. Summary statistics were provided for safety data. CDAI was evaluated by use of NRI, which considered rescued or discontinued patients as nonresponders. To determine overall long-term efficacy irrespective of rescue, analyses were also performed, in which only discontinued patients were considered as nonresponders and the observed data collected after rescue were included. The effect of reintroducing baricitinib 4 mg was evaluated in rescued patients.
RESULTS
Demographics and clinical characteristics. Supplementary Figure S3 (available with the online version of this article) shows the characteristics of 996 patients who entered the RA-BEYOND study and were rerandomized in the step-down substudy up to week 96 postrandomization. Most patients completed the full 96 weeks, with discontinuation rates of 12% and 14% for the baricitinib 4-mg and 2-mg arms, respectively. At the time of step-down randomization, demographics, and clinical characteristics—such as CRP, ESR, CDAI, and SDAI—were similar between the treatment groups (Table 1).
Patient characteristics and step-down disease activity at baseline.
Efficacy. Among those who achieved sustained disease control with baricitinib 4 mg, most patients, ranging from 60% to 91%, retained a state of LDA or REM in both groups (continued baricitinib 4 mg and step-down 2 mg) after randomization up to week 96. The step-down 2-mg arm had significantly lower LDA rates (CDAI ≤ 10) than the continued 4-mg arm at 12, 24, 48, and 96 weeks (Figures 1A,B).
(A) CDAI state with NRI for rescue or discontinuation. (B) CDAI state with NRI for discontinuation only. a NRI used after rescue treatment or for missing data. * P ≤ 0.05; ** P ≤ 0.01; *** P ≤ 0.001: Continued baricitinib 4 mg vs step-down baricitinib 2 mg. BARI: baricitinib; CDAI: Clinical Disease Activity Index; LDA: low disease activity; NRI: nonresponder imputation; REM: remission.
The percentage of patients who retained LDA progressively decreased from week 0 to week 96 in both arms. About 70% to 80% of the patients in the continued 4-mg arm and 60% to 76% in the step-down 2-mg arm retained LDA at week 96 (Figures 1A,B). However, the percentage of patients who retained REM remained relatively close to baseline 38% over time (Figures 1A,B).
Additional analyses were performed to determine overall efficacy, irrespective of rescue. Results showed that more patients achieved LDA or REM when postrescue data were used in the analyses than when postrescue data were censored and imputed as nonresponse (Figure 1B). In total, at least 76% of patients tapered to 2 mg maintained LDA/REM or recaptured with return to 4 mg (Figure 1B). Similar data were presented in Supplementary Figures S4A,B, S5A,B, and S6A,B (available with the online version of this article).
Compared with patients who continued receiving baricitinib 4 mg, dose reduction to 2 mg resulted in statistically significant increases in CDAI, SDAI, DAS28-hsCRP, and DAS28-ESR (Figure 2). Dose reduction to 2 mg also resulted in statistically significant increases in swollen joint count, tender joint count, physician global assessment, and HAQ-DI. Statistically significant differences between dose groups were not observed for other composite score components (pain, CRP, ESR, and patient global assessment; Supplementary Table S1 and Supplementary Figure S7, available with the online version of this article).
Change in (A) CDAI, (B) SDAI, (C) DAS28-hrCRP, and (D) DAS28-ESR from week 0 to 96 of step-down period. ** P ≤ 0.01; *** P ≤ 0.001: Continued baricitinib 4-mg vs step-down baricitinib 2-mg. CDAI: Clinical Disease Activity Index; DAS28: Disease Activity Score in 28 joints; DAS28-ESR: DAS28 based on ESR; DAS28-hsCRP: DAS28 based on hsCRP; ESR: erythrocyte sedimentation rate; hsCRP: high-sensitivity C-reactive protein; mLOCF: modified last observation carried forward; SDAI: Simplified Disease Activity Index.
csDMARD-IR patients (RA-BEAM, RA-BUILD, and RA-BALANCE combined). The results in this subgroup were consistent with results from the overall DMARD-IR subgroup. CDAI ≤ 10 and ≤ 2.8 response rates were similarly reduced after dose reduction for csDMARD-IR patients (Supplementary Figure S5, available with the online version of this article). Findings with respect to continuous measures for composite scores and their components were also consistent with those from the overall DMARD-IR patient group (Supplementary Table S2).
bDMARD-IR (RA-BEACON) patients. Potentially owing to the limited number of bDMARD-IR patients, differences between dose groups were not statistically significant; however, the lessoned response in LDA/REM following dose reduction was consistent between bDMARD-IR patients and csDMARD-IR patients (Supplementary Figure S6 and Supplementary Table S3, available with the online version of this article).
Effect of rescue. Rescue rates were 14.7% for those who continued receiving baricitinib 4 mg and 22.5% for those who stepped down to baricitinib 2 mg (Supplementary Figure S3, available with the online version of this article). Most rescued patients regained LDA—including REM—after rescue to baricitinib 4 mg (75.6% for baricitinib 2 mg to 4 mg), 24 weeks after rescue in the DMARD-IR group (Figure 3). In total, 76% of patients who tapered to 2 mg maintained LDA/REM or recaptured LDA/REM with return to 4 mg (Figure 1B; Supplementary Table S4).
Proportion of patients who achieved CDAI ≤ 10 after rescue. CDAI: Clinical Disease Activity Index.
Durability of treatment effect. The durability of treatment effect was assessed by investigating the kinetics of relapse. Dose reduction to 2 mg resulted in significantly more patients having quicker loss of step-down eligibility or rescue, compared to patients who continued receiving baricitinib 4 mg (Figures 4A,B; Supplementary Table S5, available with the online version of this article).
(A) Time to loss of step-down eligibility criteria. (B) Time to rescue. BARI: baricitinib; Evt: event; Pt: patient; SE: standard error.
Safety. From baseline through 96 weeks, incidence rates/100 patient-years (PYs) of step-down TEAEs (47.2 vs 46.9), TEAE infections (26.3 vs 23.3), serious AEs (9.6 vs 7.3), and serious infections (3.2 vs 1.6) were higher in patients continuing with baricitinib 4 mg compared to the step-down 2-mg group (Table 2). No difference in occurrence of venous thromboembolism, major adverse cardiovascular events, and cancer were observed between the rerandomized arms. Incidence rates/100 PYs of AEs leading to discontinuation were similar between groups (4.1 vs 3.6). Similar results were found in the CDAI REM subgroup (CDAI ≤ 2.8; Supplementary Table S6, available with the online version of this article).
Safety summary (weeks 0-96 of step-down period).
DISCUSSION
This study presents data from the completed LTE study, RA-BEYOND, up to 96 weeks after dose taper to 2 mg among patients who achieved sustained disease control with baricitinib 4 mg. This study is an extension of previously published sustained effects at 48 weeks.5 Due to the larger sample size (996 patients vs 559 patients) and the longer follow-up of our study (96 weeks vs 48 weeks), the data presented here are more robust compared to the previously reported outcomes.5
Results demonstrate that sustained LDA achieved with 4-mg baricitinib treatment can be maintained up to 96 weeks in the majority of patients (60%) who tapered to baricitinib 2 mg, compared to 70% of patients who continued on 4 mg (Figure 1A). However, most patients who tapered to 2 mg and needed rescue therapy (> 72%) recaptured LDA with return to 4 mg (Figure 3). In total, 76% of patients who tapered to 2 mg maintained LDA/REM or recaptured with return to 4 mg if needed. Consistency between changes in CDAI following step-down randomization and changes in other measures (ie, SDAI, DAS28-hsCRP, and DAS28-ESR; Figure 2) support that maintained efficacy following dose step-down is independent of using an instrument with an acute phase marker.
Patients who tapered to 2 mg experienced significantly lower incidence rates/100 PYs than those who remained at 4 mg of AEs (TEAE: 46.9 vs 47.2; TEAE infections: 23.3 vs 26.3; serious AEs: 7.3 vs 9.6; serious infections: 1.6 vs 3.2), therefore presenting a more favorable safety profile. It is notable that the risk of serious infection had a statistically significant reduction (P ≤ 0.05) of 50% when baricitinib was tapered to a 2-mg dose. No difference in occurrence of venous thromboembolism, major adverse cardiovascular events, and cancer were observed between the rerandomized arms.
Overall, 96-week efficacy and safety data after step-down to 2 mg are useful for clinicians who wish to taper patients who achieved LDA with 4 mg once daily, in line with the approved regulatory labels9-11 and clinical guidelines2-4 that recommend the tapering of DMARDs in patients with RA with sustained REM or satisfactory disease control. Results demonstrate that tapering to 2 mg is an acceptable and efficacious dose for many patients (60%). The maintenance rate of 60% LDA is a conservative estimate because of the use of NRI as an imputation method (ie, rescued or discontinued patients were counted as nonresponders). More importantly, for those who lost LDA and required rescue (n = 112), prior control of disease activity could be reestablished in the majority (76%) with return to 4 mg. Finally, the 2 mg dose may be desirable for some patients who are at an increased risk of serious infections, given the reduction of these events compared to the 4-mg treatment group.
This study has certain limitations. The protocol allowed the evaluation of dose reduction effects among patients achieving sustained control of disease activity after a minimum of 15 months of treatment with baricitinib 4 mg. This protocol requirement was set to provide sufficient stable exposure in the clinical trial but may not accommodate the timing of DMARD dose taper in clinical practice, which needs to take individual patient circumstances, treatment goals, and responses into consideration. Moreover, this study did not evaluate prognostic factors for successful dose tapering from 4 mg to 2 mg either prior to baricitinib treatment initiation or after LDA was achieved with treatment of baricitinib 4 mg. However, considering that ≥ 60% of patients maintained control of disease activity after taper to 2 mg and ≥ 76% of patients recaptured control of disease activity with rescue to 4 mg, the subset of patients who would not respond to taper or rescue would be relatively small. Future analyses and modeling of these factors may aid decision making when selecting individual patients starting on baricitinib 4 mg to proceed to a step-down dose that balances the benefits and risks of a 4-mg vs 2-mg dose.
These results from the completed phase III randomized dose-taper study indicate that in patients with moderate and severe RA, for whom sustained clinical disease control has been induced with baricitinib 4 mg once daily, dose taper to baricitinib 2 mg results in increased disease activity for some patients. However, the majority of patients can retain LDA following dose taper or regain it with return to 4 mg if needed. Lower incidence rates of TEAEs and serious AEs (including infections) were observed after step-down in the dose-tapered group compared with patients who continued baricitinib 4 mg. Therefore, dose tapering to 2 mg baricitinib after induction of sustained RA control with 4 mg appears reasonable. However, the decision to dose taper should be part of the shared decision-making process between the patient and physician, considering the balance between risk and benefit.
ACKNOWLEDGMENT
The authoring team would like to thank Lynn Buckley for medical writing support, and Jens Gammeltoft Gerwien, Samuel Ogwu, and Anthony Marca for their support in providing medical, statistical, and editorial reviews, respectively.
Footnotes
CONTRIBUTIONS
All authors participated in data analysis and interpretation, reviewing of drafts and the final manuscript, and critical revision (review and editing) of the work. DM, EH, PF, KH, YT, and MT also contributed to conception, design, acquisition, and analysis. The authors vouch for the veracity and completeness of the data and data analyses.
FUNDING
This study was funded by Eli Lilly and Company.
COMPETING INTERESTS
CJE received clinical grants and support for attending meetings and/or travel from Eli Lilly and Company; and payment of honoraria for lecture fees and consultancy related to the subject area covered in the manuscript from AbbVie, Alpha Sigma, BMS, Fresenius, Galapagos, Gilead, Janssen, Lilly, Pfizer, MSD, Novartis, Roche, Sanofi, and UCB. GK received consulting fees, payment for lectures, and support for attending meetings and/or travel from Eli Lilly and Company. JA has received grants or contracts from Pfizer, BMS, Novartis, Galapagos/Alfasigma, and Nordic; consulting fees from AbbVie, Sanofi, and Nordic; payment or honoraria for lectures, presentations, and/or speakers bureaus from Galapagos, AbbVie, BMS, Sanofi, Roche-Chugai, Nordic, Biogen, Fresenius Kabi, MSD, Celltrion, Medac, and Novartis; support for attending meetings and/or travel from Celltrion, AbbVie, and Novartis; and participated on a data safety monitoring board or advisory board for AbbVie, Pfizer, Galapagos/Alfasigma, and BMS. FC has received clinical grants, speaker fees, and support for attending meetings and/or travel from Eli Lilly and Company; and speaker fees from Eli Lilly and Company, Biogen, BMS, Galapagos, AbbVie, Pfizer, and UCB. AB has received support for the present manuscript and for attending meetings and/or travel from Eli Lilly and Company; grants or contracts from AbbVie, UCB, and Pfizer; consulting fees from Eli Lilly and Company, UCB, AbbVie, Pfizer, Galapagos, Sanofi, Nordic, Sandoz, Gebro, BMS; payment or honoraria for lectures, presentations, and/or speakers bureaus from Eli Lilly and Company, UCB, AbbVie, Pfizer, Galapagos, Sanofi, Nordic, Sandoz, Gebro, and BMS; and participated in a data safety monitoring board or advisory board for Eli Lilly and Company and UCB. DM, EH, and PF are employees and shareholders of Eli Lilly and Company. MT, YT, and KH are employees of Eli Lilly Japan K.K. and shareholders of Eli Lilly and Company. TT has received consulting fees from AbbVie, Mitsubishi-Tanabe, Eli Lilly Japan K.K., Taisho, and Gilead; payment or honoraria for lectures, presentations, and/or speakers bureaus from AstraZeneca K.K., Astellas, AbbVie, Mitsubishi-Tanabe, Pfizer Japan, Chugai, Eisai, Eli Lilly Japan K.K., Janssen Pharma K.K., Taisho, and Gilead. ZL reports no conflicts of interest relevant to this article.
ETHICS AND PATIENT CONSENT
All patients provided written informed consent before the first study procedure in RA-BEYOND. Consent for participation in the step-down substudy was obtained at the time of entry into RA-BEYOND. Lilly or its representatives provided data, laboratory, and site monitoring services. The study was designed by the sponsor, Eli Lilly and Company, an academic advisory board including non-Lilly authors of this manuscript, and Incyte. It was conducted in accordance with the ethical principles of the Declaration of Helsinki and Good Clinical Practice guidelines.
- Accepted for publication December 12, 2024.
- Copyright © 2025 by the Journal of Rheumatology
This is an Open Access article, which permits use, distribution, and reproduction, without modification, provided the original article is correctly cited and is not used for commercial purposes.
REFERENCES
SUPPLEMENTARY DATA
Supplementary material accompanies the online version of this article.
