Abstract
Objective. To evaluate the effect of tofacitinib (TOF) on American College of Rheumatology (ACR) response criteria components in patients with rheumatoid arthritis (RA).
Methods. This post hoc analysis pooled data from RA phase III randomized controlled trials (RCTs) assessing TOF 5 or 10 mg BID, adalimumab (ADA), or placebo, with conventional synthetic disease-modifying antirheumatic drugs, and a phase IIIb/IV RCT assessing TOF 5 mg BID monotherapy, TOF 5 mg BID with methotrexate (MTX), or ADA with MTX. Outcomes included proportions of patients achieving ACR20/50/70 responses and ≥ 20/50/70% improvement rates in ACR components at week 2 and months 1, 3, and 6; and mean percent improvement in ACR components and Clinical or Simplified Disease Activity Index (CDAI or SDAI) low disease activity or remission rates, at month 3, for ACR20/50/70 responders.
Results. Across treatment groups, ≥ 20/50/70% improvement rates were numerically higher for most physician- vs patient-reported measures. In phase III RCTs, at earlier timepoints, ≥ 50/70% improvements in patient global assessment of disease activity, pain, and physician global assessment were similar. Among ACR20 responders receiving TOF, mean percent improvements for tender and swollen joint counts were > 70% at month 3. CDAI/SDAI remission was achieved at month 3 by 27.8–45.0% of ACR70 responders receiving TOF.
Conclusion. Among ACR20 responders treated with TOF, physician-reported components particularly exceeded 20% response improvement. At month 3, disease state generally did not corroborate ACR70 response criteria. Divergences between physician- and patient-reported measures highlight the importance of identifying appropriate patient-reported outcome targets to manage RA symptoms in clinical practice. (ClinicalTrials.gov: NCT00847613/NCT00856544/NCT00853385/NCT02187055)
Composite measures of disease activity and treatment response are key efficacy outcomes in rheumatoid arthritis (RA) clinical trials. Common measures include the American College of Rheumatology (ACR) response criteria,1 which require meeting a threshold of ≥ 20/50/70% improvement (ACR20/50/70, respectively) in several physician- and patient-reported measures, including tender and swollen joint counts (TJC and SJC, respectively; primary criteria) and at least 3 of 5 secondary criteria: physician global assessment (PGA), patient global assessment of disease activity (PtGA), patient-reported pain (Pain), Health Assessment Questionnaire–Disability Index (HAQ-DI), and C-reactive protein (CRP).2,3 The ACR20 response rate is commonly used as the primary outcome measure in RA trials because it can discriminate between active treatment and placebo.1,3 However, ACR response outcomes are not typically calculated in real-world practice.4
Other composite measures assessed in RA trials and the clinic include the Simplified Disease Activity Index (SDAI; numerical sum of TJC, SJC, PGA, PtGA, and CRP),5 and the Clinical Disease Activity Index (CDAI; numerical sum of TJC, SJC, PGA, and PtGA).6 Cut-offs of both indices have been established to classify disease state (remission or low, moderate, or high disease activity) in patients with RA.7
Using a treat-to-target approach for RA, physicians are recommended to tailor treatment plans to prespecified goals (eg, remission or low disease activity [LDA]), and to use composite disease activity measures to monitor treatment response.8 However, there is no standard composite measure used across clinics. Therefore, a comprehensive understanding of the effect of RA treatments on the individual components, which are typically assessed in the clinic, and the resulting impact on composite measures, may help inform physicians regarding a patient’s response to treatment.
Tofacitinib (TOF) is an oral Janus kinase (JAK) inhibitor for the treatment of RA. The efficacy and safety of TOF immediate-release 5 and 10 mg BID, administered as monotherapy or in combination with conventional synthetic disease-modifying antirheumatic drugs (csDMARDs), predominantly methotrexate (MTX), in patients with moderately to severely active RA, have been previously reported in phase II,9,10,11,12,13 phase III,14–20 and phase IIIb/IV21 randomized controlled trials (RCTs) of up to 24 months’ duration and in long-term extension (LTE) studies with up to 114 months’ observation.22,23,24 The long-term safety of TOF has been reported in an integrated safety analysis of RCTs and LTE studies spanning 114 months’ cumulative TOF exposure.25
To provide further insight for clinicians regarding expected outcomes with TOF, we examined the effect of TOF 5 or 10 mg BID, adalimumab (ADA) 40 mg every 2 weeks, or placebo, with background csDMARDs, or TOF 5 mg BID monotherapy, on the ACR components through a post hoc analysis of phase III and phase IIIb/IV trials. To further explore the ACR components that are most (and least) likely to be improved by treatment, and meet the improvement thresholds (≥ 20/50/70%) required for inclusion in the ACR response calculation, we evaluated the relative contribution of the secondary ACR components (PGA, PtGA, Pain, HAQ-DI, and CRP) to the overall ACR20/50/70 response rates. Further, to provide insight into expected disease state outcomes in patients achieving ACR20/50/70 responses in clinical trials, we show the proportions of responders achieving SDAI- and CDAI-defined LDA and remission.
METHODS
Study design and patients. This post hoc analysis included 2 cohorts. The placebo-controlled cohort comprised pooled data from 3 phase III RCTs (ORAL Scan [ClinicalTrials.gov: NCT00847613], ORAL Standard [ClinicalTrials.gov: NCT00853385], and ORAL Sync [ClinicalTrials.gov: NCT00856544]) of TOF in patients with active RA. The head-to-head cohort comprised data from a phase IIIb/IV RCT (ORAL Strategy [ClinicalTrials.gov: NCT02187055]) of TOF vs ADA in patients with active RA. Study designs and patient inclusion/exclusion criteria were reported previously16,17,19,20,21 and are summarized in Table 1.
Each study was conducted in accordance with the Declaration of Helsinki, International Conference on Harmonization Guidelines for Good Clinical Practice, and local country regulations, and was approved by the institutional review board and independent ethics committee at each center.16,17,19,20,21 Patients provided written informed consent. No further ethical approval was required for this post hoc analysis in accordance with the policies of our institutions.
Outcomes. Outcomes assessed included the following: proportions of patients achieving overall ACR20/50/70 responses and ≥ 20/50/70% improvements from baseline in ACR components (improvement rates; TJC, SJC, PGA, PtGA, Pain, HAQ-DI, and CRP) at week 2 (ORAL Sync only), week 6 (head-to-head cohort only), month 1 (placebo-controlled cohort only), month 3 (all studies; end of placebo-controlled period for placebo-controlled cohort), and month 6 (all studies); the relative contribution of the secondary ACR components (PGA, PtGA, Pain, HAQ-DI, and CRP) to overall ACR20/50/70 response rates at month 3; and the relative contribution of SDAI components (TJC, SJC, PGA, PtGA, and CRP) and CDAI components (TJC, SJC, PGA, and PtGA) to mean change from baseline in SDAI and CDAI scores, respectively, at month 3.
Additional outcomes assessed in subgroups of patients achieving ACR20/50/70 responses (ACR20/50/70 responders) at month 3 included the following: mean percent improvement from baseline in ACR components, SDAI score, and CDAI score; and proportions of patients achieving LDA and remission as defined by SDAI (≤ 11 and ≤ 3.3, respectively) and CDAI (≤ 10 and ≤ 2.8, respectively).7
Statistical analyses. Data are presented for the full analysis set, which comprised all patients who were randomized and received ≥ 1 dose of study treatment. Analyses are based on observed case data in patients with all 7 ACR components assessed at the analyzed timepoint. No imputation was performed for missing data. Outcomes were summarized descriptively and numerical differences (with no formal statistical comparisons) between treatments are reported.
To assess the relative contribution of each secondary ACR component to the attainment of the overall ACR20 response rate, each component was sequentially set to “no improvement” (ie, value of 0 in change from baseline) and the ACR20 response rate was recalculated. The resulting response rates were then rank-ordered from 1–5, with 1 representing the largest contribution (largest decrease in ACR20 response rate); and 5 representing the smallest contribution (smallest decrease in ACR20 response rate). This approach was used to assess the relative contribution of each secondary ACR component to the attainment of the overall ACR50 and ACR70 response rates. The relative contributions of TJC and SJC to the attainment of ACR20/50/70 response rates were not considered in this analysis as these are primary components, with ≥ 20/50/70% improvement in these components required to achieve an ACR20/50/70 response rate, respectively. The relative contribution of each SDAI and CDAI component to the mean change from baseline in SDAI and CDAI, respectively, was assessed using a similar approach.
RESULTS
Patients. The placebo-controlled cohort comprised 2117 patients receiving TOF 5 mg BID + csDMARDs (n = 769), TOF 10 mg BID + csDMARDs (n = 767), ADA + csDMARDs (n = 191), or placebo + csDMARDs (n = 390). The head-to-head cohort comprised 1088 patients receiving TOF 5 mg BID monotherapy (n = 359), TOF 5 mg BID + MTX (n = 363), or ADA + MTX (n = 366). Demographics and baseline characteristics for each study were previously reported.16,17,19,21
ACR20/50/70 response rates and ≥ 20/50/70% improvement rates in ACR components. In the placebo-controlled cohort, ACR20/50/70 response rates and ≥ 20/50/70% improvement rates in ACR components at month 3 (end of placebo-controlled period) were similar with TOF 5 or 10 mg BID + csDMARDs and ADA + csDMARDs, but were higher with active treatment vs placebo (Figure 1A–C). Notably, ACR20/50/70 response rates and ≥ 20/50/70% improvement rates in ACR components were higher beginning at week 2 and month 1 with TOF and ADA, respectively, vs placebo.
Across treatments, ACR20/50/70 response rates and ≥ 20/50/70% improvement rates in ACR components mostly increased through month 6, and improvement rates for most ACR components surpassed ACR20/50/70 response rates (Figure 1A–C). Typically, through month 6, ≥ 20/50/70% improvement rates were higher in physician-reported measures (TJC, SJC, and PGA) vs patient-reported measures (PtGA, Pain, and HAQ-DI) across treatments. However, ≥ 20/50/70% improvement rates in CRP were highest vs the other components at week 2 with active treatment and tended to remain stable over time. Differences between improvement rates in PGA vs the patient-reported measures were generally greater when considering ≥ 20% vs ≥ 50/70% improvement rates through month 6 across treatments, particularly at earlier timepoints; indeed, ≥ 50/70% improvement rates in PtGA and Pain were similar to PGA at week 2, month 1, and month 3 (≥ 70% improvement rates only). Generally, ≥ 20/50/70% improvement rates in PtGA and Pain were comparable and followed a similar pattern through month 6, regardless of treatment. Across treatments, ≥ 20/50/70% improvement rates in HAQ-DI showed the least improvement of all ACR components through month 6.
In the head-to-head cohort, ACR20/50/70 response rates and ≥ 20/50/70% improvement rates in ACR components were comparable through month 6 in patients receiving TOF 5 mg BID monotherapy, TOF 5 mg BID + MTX, or ADA + MTX (Supplementary Figure 1A–C, available with the online version of this article). Across treatments, ≥ 20/50/70% improvement rates for most ACR components surpassed ACR20/50/70 response rates; however, ≥ 20/50/70% improvement rates for HAQ-DI were generally similar to or lower than the ACR20/50/70 response rates.
Similar to the placebo-controlled cohort, ≥ 20/50/70% improvement rates in ACR components were higher in physician-reported vs patient-reported measures. Across treatments, ≥ 20/50/70% improvement rates in PtGA and Pain were typically comparable and followed a similar pattern through month 6. Unlike the placebo-controlled cohort, ≥ 50/70% improvement rates in PtGA and Pain were not similar to PGA through month 6.
Relative contribution of the secondary ACR components to ACR20/50/70 response rates at month 3. Generally, in the placebo-controlled cohort, PGA contributed most, whereas HAQ-DI contributed least, to ACR20/50/70 response rates at month 3, across treatments (Table 2A–C). Exceptions with active treatment included ACR70 response rates with TOF 5 mg BID + csDMARDs, wherein Pain contributed most, ACR50 response rates with TOF 10 mg BID + csDMARDs, wherein PtGA contributed least, and ACR70 response rates with ADA + csDMARDs, wherein PGA and PtGA contributed most and Pain and CRP contributed least (Table 2B–C).
In the head-to-head cohort, results aligned with the placebo-controlled cohort, as PGA generally contributed most, whereas HAQ-DI generally contributed least, to ACR20/50/70 response rates at month 3, across treatments (Supplementary Table 1A–C, available with the online version of this article).
Percent improvement from baseline in ACR components, SDAI score, and CDAI score in ACR20/50/70 responders at month 3. Among ACR20/50/70 responders in the placebo-controlled cohort, mean percent improvements from baseline in ACR components typically exceeded 20/50/70%, respectively, at month 3, across treatments (Figure 2A–C). Mean percent improvement from baseline was higher for TJC and SJC vs PGA, PtGA, Pain, HAQ-DI, and CRP in ACR20/50/70 responders across treatments, and greater differences were observed between physician-reported and patient-reported measures in ACR20 vs ACR50/70 responders. Mean percent improvement from baseline in Pain was comparable in ACR20/50 responders receiving TOF vs ADA; however, mean percent improvement from baseline in Pain was 10.4% higher in ACR70 responders receiving TOF 5 mg BID + csDMARDs vs ADA + csDMARDs. Mean percent improvements from baseline in CRP appeared lower with TOF vs ADA because of outliers in the TOF groups; median percent improvements from baseline in CRP were similar across active treatment groups for ACR20 (73.8–81.6%), ACR50 (76.1–85.3%), and ACR70 responders (75.9–85.9%). Notably, for ACR20 responders receiving TOF, mean percent improvement from baseline in TJC and SJC exceeded 70%.
Across ACR20/50/70 responders, mean percent improvements from baseline in SDAI and CDAI scores to month 3 were generally similar to improvements observed for TJC and SJC with all treatments, and exceeded 65% in ACR20 responders receiving TOF (Figure 2A–C).
In the head-to-head cohort, mean percent improvements in ACR components, SDAI score, and CDAI score in ACR20/50/70 responders at month 3 were comparable across treatments and mostly similar to the placebo-controlled cohort (Supplementary Figure 2A–C, available with the online version of this article).
Proportions of ACR20/50/70 responders achieving SDAI- or CDAI-defined LDA or remission at month 3. In the placebo-controlled cohort, the proportions of ACR20/50/70 responders achieving SDAI- or CDAI-defined LDA or remission at month 3 were higher with active treatment vs placebo, with the largest differences observed in ACR70 responders receiving active treatment vs placebo (Figure 3). At month 3, the proportions of ACR20/50/70 responders in the active treatment groups achieving CDAI LDA ranged from 38.2–49.1%, 61.4–73.6%, and 88.2–95.0%, respectively, vs 28.8%, 60.6%, and 62.5%, respectively, for placebo; those achieving CDAI remission ranged from 3.6–9.0%, 9.1–16.7%, and 23.5–45.0%, respectively, vs 0.9%, 3.0%, and 12.5%, respectively, for placebo. Overall, SDAI and CDAI LDA and remission rates were numerically higher in ACR20/50/70 responders receiving TOF vs ADA.
In the head-to-head cohort, the proportions of ACR20/50/70 responders across treatments achieving CDAI LDA at month 3 were mostly comparable with the placebo-controlled cohort (Supplementary Figure 3, available with the online version of this article). The proportions of ACR20/50/70 responders across treatments achieving CDAI remission at month 3 were generally higher than the placebo-controlled cohort, ranging from 10.0–10.8%, 17.9–18.8%, and 30.6–40.0%, respectively. Across treatments, SDAI and CDAI LDA and remission rates were generally similar (Supplementary Figure 3).
Relative contribution of SDAI or CDAI components to mean change from baseline in SDAI or CDAI scores at month 3. In the placebo-controlled cohort, TJC, followed by SJC, contributed most to mean change from baseline in SDAI at month 3, whereas CRP contributed least, across treatments (Table 3). Similarly, across treatments, TJC, followed by SJC, contributed most to mean change from baseline in CDAI at month 3, whereas PtGA contributed least (Table 3). Similar results were observed in the head-to-head cohort (Supplementary Table 2, available with the online version of this article).
DISCUSSION
Many RA clinical trials use ACR response criteria, generally ACR20/50, as the primary outcome measure, whereas only the individual ACR components are commonly assessed in clinical practice.1,26 Therefore, to guide clinical decision making, it is important for physicians to have a comprehensive understanding of the effect of RA treatments on the composite measures and individual components. This post hoc analysis investigated the effect of treatments in phase III and phase IIIb/IV TOF RA RCTs on ACR components, SDAI score, and CDAI score; the relative contribution of each component to the ACR20/50/70 response rates, SDAI score, and CDAI score; and the proportions of patients achieving SDAI and CDAI LDA and remission stratified by ACR20/50/70 response.
In the placebo-controlled cohort, ACR20/50/70 response rates and ≥ 20/50/70% improvement rates in ACR components at month 3 were higher with active treatment vs placebo, aligning with previous findings.16,17,19 These improvements were observed from the earliest timepoint for TOF-treated and ADA-treated patients (week 2 and month 1, respectively), vs placebo-treated patients. In the head-to-head cohort, ACR20/50/70 response rates and ≥ 20/50/70% improvement rates in ACR components were typically similar across timepoints and treatments. In ACR20/50/70 responders in both cohorts, mean percent improvements from baseline in the ACR components typically exceeded 20/50/70%, respectively, at month 3. These findings provide insight into the impact that TOF or ADA have on the ACR components at month 3, a key timepoint for treatment target assessments, according to the treat-to-target approach for RA.8
Generally, in both cohorts, ≥ 20/50/70% improvement rates were higher in physician-reported vs patient-reported measures through month 6. However, in the placebo-controlled cohort, ≥ 20/50/70% improvement rates in CRP were highest vs the other components at the earliest timepoint for TOF-treated and ADA-treated patients (week 2 and month 1, respectively) and tended to remain consistent over time. This corresponds with prior findings that TOF results in a rapid, early reduction in CRP that stabilizes by month 1.27,28 Overall, however, mean percent improvements from baseline in CRP appeared to be lower in TOF-treated patients compared with ADA-treated patients. This was likely because of outliers in the TOF groups, in which the majority of patients in this analysis were randomized, as median improvements from baseline were similar across treatments. However, there remains a possibility that this could also be reflective of mechanistic differences between JAK inhibitors and tumor necrosis factor inhibitors in the modulation of inflammatory mediators. Interestingly, ≥ 70% improvement rates in PGA, PtGA, and Pain were similar through month 3 in the placebo-controlled cohort, suggesting that patient and physician perceptions of disease status may be more closely aligned for patients achieving LDA following RA treatment. These results may also provide insight into the impact of TOF on other composite measures assessed in RA trials and clinics, including Disease Activity Score in 28 joints using erythrocyte sedimentation rate,29 which was previously reported in TOF-treated vs placebo-treated patients.16,17,19,20
In both cohorts, physician-reported measures typically contributed more to ACR20/50/70 response rates, SDAI score, and CDAI score than patient-reported measures at month 3. This may be because of differences in the type of information collected between physician-reported and patient-reported measures,30 or discrepancies in disease or symptom perceptions between the physician and patient, as reported previously for RA31,32,33,34 and other inflammatory diseases.35 Additionally, the divergence of ≥ 50/70% improvement rates in PGA vs PtGA and Pain at later timepoints in the placebo-controlled cohort may highlight an unintentional overestimation of treatment response by physicians who wish to retain patients in clinical trials despite only achieving mild symptomatic resolution. Interestingly, differences in ≥ 20/50/70% improvement rates in PGA vs PtGA and Pain were observed across most timepoints and treatments in the head-to-head cohort, potentially because of physicians’ higher expectations of improvement in the absence of placebo. These results further highlight the need to identify appropriate patient-reported outcome targets that help assess RA symptoms in clinical practice. In both cohorts, HAQ-DI generally contributed least to the achievement of ACR20/50/70 response rates, which may be attributed to preexisting functional restrictions of the patients in these trials. While these results may indicate that HAQ-DI is not a reliable measure of treatment response, it is unclear how these results translate into clinical practice. However, a previous study showed no statistical differences in HAQ-DI between RCTs and observational studies.36 Future analyses could focus on an earlier RA patient population to identify if HAQ-DI contributes more to the achievement of ACR response rates in those patients vs the cohorts studied here.
Analysis of ACR20/50/70 responders achieving SDAI- or CDAI-defined LDA or remission at month 3 showed that 38.2–52.0% of ACR20 responders and 61.4–79.5% of ACR50 responders receiving active treatment across both cohorts achieved SDAI or CDAI LDA. Further, while many physicians consider the ACR70 response rate to correspond with a state of remission, these results showed that 23.5–45.0% of ACR70 responders receiving active treatment across both cohorts achieved SDAI or CDAI remission. This suggests that ACR response rates, which measure treatment response between 2 timepoints,37 may not corroborate achievement of disease state thresholds when assessing the effectiveness of RA treatments in a clinical setting.
Accumulating evidence indicates that JAK inhibitors may be more effective in improving patient pain than other advanced therapies.38,39 This analysis showed that ≥ 20/50/70% improvement rates in Pain were generally comparable in TOF-treated and ADA-treated patients in both cohorts, as previously reported.40 However, there was a small improvement in Pain in ACR70 responders receiving TOF vs ADA at month 3. Therefore, a specific effect on pain, other than through the established antirheumatic efficacy of TOF, cannot be concluded from this analysis. The finding that ≥ 20/50/70% improvement rates in PtGA and Pain were similar across treatments through month 6 confirm previous results over 3 months, and identify a close concordance of PtGA and Pain outcomes.41 Physicians base their overall assessment of patients’ disease activity on joint counts (TJC and SJC), whereas PtGA is greatly influenced by noninflammatory patient factors,42 which might explain the lower improvement rate in PtGA vs PGA.
Limitations include the post hoc nature of this evaluation, the lack of formal statistical testing, and the small sample size for ACR70 responders. In the placebo-controlled cohort, comparisons between active treatments were limited, as data for TOF were pooled across 3 studies vs 1 study for ADA. Further, responses in the TOF groups at month 6 may have been affected by the advancement of nonresponder patients to TOF at month 3. This analysis was conducted over 6 months; further studies would be required to determine the effect of long-term treatment and disease status on the components of the composite measures. Finally, interpretation of these data in a real-world context is limited by the clinical trial setting.
In conclusion, this post hoc analysis of data from phase III and phase IIIb/IV TOF RA RCTs provides insight into patient responses to TOF or ADA in terms of the composite measures and their individual components. In particular, the higher weighting of physician-reported vs patient-reported components to overall response and remission highlights the importance of considering the patient’s perspective when making treatment decisions. Despite different modes of action, TOF and ADA did not show differentiated efficacy across the composite measures studied. Although ACR20 response rate is typically thought of as a “low” threshold to meet, these findings demonstrate that in patients meeting composite score criteria, TOF-induced improvements in clinically measured signs and symptoms, especially those reported by the physician, far exceed what would be considered a 20% improvement in response. Last, these data show that changes in disease state do not generally corroborate ACR70 response criteria. Taken together, these findings may help clinicians to interpret clinical study results, and to define expected responses to advanced therapies, to assist in setting treatment goals for patients during routine practice.
ACKNOWLEDGMENT
The authors would like to acknowledge Tanya Girard, PhD, a former employee and stockholder of Pfizer Canada ULC, for her significant intellectual and scientific contributions to this work. Medical writing support, under the guidance of the authors, was provided by Jennifer Arnold, PhD, CMC Connect, McCann Health Medical Communications and was funded by Pfizer Inc, New York, New York, USA, in accordance with Good Publication Practice (GPP3) guidelines. Some data reported in this manuscript were previously presented at the American College of Rheumatology Annual Scientific Meeting, Atlanta, Georgia, USA, November 8–13, 2019: Bessette L, Dougados M, Mysler E, et al. Impact of tofacitinib on the individual components of the ACR composite score in patients with rheumatoid arthritis: a post hoc analysis of phase 3 trials [abstract]. Arthritis Rheumatol 2019;71 Suppl 10:1348.
Footnotes
This study was sponsored by Pfizer Inc.
LB has received grants and/or research support from AbbVie, Amgen, BMS, Celgene, Janssen, Eli Lilly, Gilead Sciences, Novartis, Pfizer Inc, Roche, Sanofi, and UCB; and has acted as a consultant or speaker for AbbVie, Celgene, Eli Lilly, Fresenius Kabi, Gilead Sciences, Novartis, Pfizer Inc, and Sandoz. EM has received research grants from Eli Lilly, Pfizer Inc, and Roche; and serves on speakers’ bureaus for AbbVie, Amgen, AstraZeneca, BMS, Eli Lilly, Janssen, Pfizer Inc, Roche, and Sanofi. CDK, KK, TL, and PVO are employees and stockholders of Pfizer Inc. RFvV has received grants and/or research support from AbbVie, Amgen, BMS, GSK, Pfizer Inc, Roche, and UCB; and has acted as a consultant for AbbVie, AstraZeneca, Biotest, BMS, Celgene, Crescendo, Eli Lilly, GSK, Janssen, Merck, Novartis, Pfizer Inc, Roche, UCB, and Vertex.
- Accepted for publication February 9, 2022.
- Copyright © 2022 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
DATA AVAILABILITY
Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions, and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.