Effect of Tocilizumab on Disease Activity in Patients With Active Polymyalgia Rheumatica Receiving Glucocorticoid Therapy: A Randomized Clinical Trial

Key Points

Question  Among patients with active polymyalgia rheumatica despite glucocorticoid therapy, does the addition of tocilizumab, compared with placebo, result in lower disease activity and lower glucocorticoid dosage?

Findings  In this randomized clinical trial that included 100 patients, a low disease activity score (polymyalgia rheumatica activity score computed using the C-reactive protein level <10) with a reduction in prednisone dose to less than or equal to 5 mg per day or a decrease greater than or equal to 10 mg per day at 24 weeks was achieved in 63.7% of patients in the tocilizumab group and 31.4% of patients in the placebo group, which was a statistically significant difference.

Meaning  Among patients with active polymyalgia rheumatica receiving glucocorticoid therapy, the addition of tocilizumab, compared with placebo, resulted in improved disease activity scores with reduced prednisone dosage at 24 weeks, although additional research is needed to confirm efficacy and assess the balance of potential benefits and harms.

Importance  Few treatments are available for patients with glucocorticoid-dependent polymyalgia rheumatica. IL-6 antagonists may reduce disease activity in patients with active glucocorticoid-dependent polymyalgia rheumatica.

Objective  To compare the efficacy of tocilizumab vs placebo in patients with glucocorticoid-dependent polymyalgia rheumatica.

Design, Setting, and Participants  This double-blind, parallel-group, placebo-controlled randomized clinical trial enrolled 101 patients with polymyalgia rheumatica at 17 hospitals in France from February 2017 to October 2019. Final follow-up occurred in November 2020. Inclusion criteria were persistent disease activity (polymyalgia rheumatica activity score computed using the C-reactive protein level [CRP PMR-AS] >10) and prednisone dose greater than or equal to 10 mg per day.

Interventions  Patients were randomly assigned to receive intravenous tocilizumab (8 mg/kg; n = 51) or placebo (n = 50) every 4 weeks for 24 weeks, combined with predefined standardized tapering of oral prednisone.

Main Outcomes and Measures  The primary efficacy end point was CRP PMR-AS less than 10 (range, 0-100; higher values indicate greater activity; no minimal clinically important difference defined) combined with either prednisone dose less than or equal to 5 mg per day or a decrease in prednisone dose greater than or equal to 10 mg from baseline at week 24. There were 11 secondary outcomes assessed at week 24 included in this report, including disease activity (measured by CRP PMR-AS) and the proportion of patients no longer taking prednisone.

Results  Of the 101 randomized patients (mean age, 67.2 years; 68 [67.3%] women), 100 (99%) received at least 1 infusion and 100 completed the trial. The primary end point was achieved in 67.3% of patients in the tocilizumab group and 31.4% of patients in the placebo group (adjusted difference, 36.0% [95% CI, 19.4%-52.6%]; adjusted relative risk, 2.3 [95% CI, 1.5-3.6]; P < .001). Of 11 reported secondary end points at 24 weeks, 7 showed significant differences favoring tocilizumab, including mean CRP PMR-AS score (7.5 [95% CI, 5.4-9.6] vs 14.9 [95% CI, 11.4-18.4]; adjusted difference, −7.5 [95% CI, −11.2 to −3.8]; P < .001) and the percentage of patients no longer receiving prednisone (49.0% vs 19.6%; adjusted difference, 29.3% [95% CI, 18.9%-39.7%]; adjusted relative risk, 2.5 [95% CI, 1.8-3.5]; P < .001). The most frequent adverse events were infections, experienced by 23 patients (46.9%) in the tocilizumab group and 20 (39.2%) in the placebo group.

Conclusions and Relevance  Among patients with active polymyalgia rheumatica despite prednisone therapy, tocilizumab, compared with placebo, resulted in a significantly greater percentage of patients with a CRP PMR-AS less than 10 with reduced prednisone requirements at week 24. Further research is needed to confirm efficacy and to determine the balance of potential benefits and harms.

Trial Registration  ClinicalTrials.gov Identifier: NCT02908217

Based on data from 2016, the lifetime risk of polymyalgia rheumatica is 2.4% for women and 1.7% for men, with the incidence increasing between ages 50 and 80 years.¹ Polymyalgia rheumatica is associated with pain and stiffness of the shoulder and pelvic girdles, with elevated acute phase reactants.¹

Glucocorticoids are the primary treatment for polymyalgia rheumatica, although approximately² 50% of patients relapse during the first year and 25% require treatment for approximately 4 to 8 years.^3-5 Alternative effective therapeutic strategies that avoid adverse effects associated with long-term glucocorticoid prescriptions are needed.^5,6

The pathogenesis of polymyalgia rheumatica is unclear.^7,8 The cytokine IL-6 appears to be involved in the pathology of polymyalgia rheumatica. The humanized anti–IL-6 receptor tocilizumab has been approved for giant cell arteritis.⁹ Three prospective open-label phase 2 studies of polymyalgia rheumatica suggested that tocilizumab might improve disease activity and diminish glucocorticoid requirements.^10-12 For example, in an open-label study of 20 patients with newly diagnosed polymyalgia rheumatica without prior glucocorticoid therapy,^10,13 tocilizumab treatment was associated with a polymyalgia rheumatica activity score (PMR-AS) less than or equal to 10 at week 12 among all included patients.¹⁴

Therefore, the current randomized, double-blind, parallel-group, placebo-controlled clinical trial, SEMAPHORE (Safety and Efficacy of Tocilizumab vs Placebo in Polymyalgia Rheumatica With Glucocorticoid Dependence), was designed to determine whether tocilizumab, compared with placebo, increased the frequency of low disease activity while reducing the dosage of prednisone therapy at week 24.

This randomized, double-blind, phase 3 clinical trial compared intravenous infusions of tocilizumab or placebo in patients with polymyalgia rheumatica treated at 17 hospitals in France. The trial was conducted in accordance with the Declaration of Helsinki and International Conference on Harmonization Good Clinical Practice Guideline. Recruitment occurred between February 2017 and October 2019. The trial protocol was approved by the appropriate ethics committee (Comités de Protection des Personnes Ouest VI). Patients provided written informed consent. An independent data and safety monitoring board provided oversight. The protocol and statistical analysis plan are provided in Supplements 1 and 2.

Eligible participants were older than 50 years and met criteria for Chuang’s polymyalgia rheumatica classification.¹⁵ Participants were eligible if they had a C-reactive protein (CRP) level greater than or equal to 10 mg/L or erythrocyte sedimentation rate (ESR) greater than or equal to 20 mm per hour at disease onset; responded to 12 to 25 mg per day of prednisone (attained CRP level less than 10 mg/L or ESR less than 20 mm per hour while receiving glucocorticoids); and subsequently developed glucocorticoid dependency, defined as PMR-AS calculated using CRP (CRP PMR-AS) greater than 10 during prednisone tapering according to international recommendations when prednisone dosage lower than 10 mg per day was reached.² Concomitant methotrexate or hydroxychloroquine treatment in a stable dose for at least the past 3 months, with no plans to change this dose, did not preclude inclusion.

Exclusion criteria (see eTable 4 in Supplement 3) included symptoms suggesting giant cell arteritis; uncontrolled dyslipidemia; unstable hypertension or cardiovascular disease; history of cancer; severe or unstable acute or chronic disease; manifestations of another inflammatory rheumatic or connective tissue disease or hydroxyapatite crystal or pyrophosphate crystal disease; severe hip and/or shoulder osteoarthritis; planned surgical procedure; investigator discretion based on laboratory abnormalities or other health conditions; current antibiotic therapy; drug or alcohol abuse; hepatitis B or C infection, HIV infection, or tuberculosis; active thyroid disease; and drug-related myalgia. Other exclusion criteria were prednisone dosage greater than 25 mg per day, immunosuppressant therapy other than methotrexate, absence of effective contraception in patients able to conceive, and inability to understand the study protocol.

Patients were randomized in a 1:1 ratio by the Brest University Hospital (study sponsor) to receive intravenous infusions of tocilizumab or placebo. Randomization was stratified by hospital and baseline CRP PMR-AS (>17 or ≤17) using the Ennov Clinical Interactive Web Response System. The randomization sequence was generated by the study sponsor using random block sizes of 2, 4, or 6. Electronic case report forms were used to obtain the randomization number for each patient. Investigators, patients, outcome assessors, and sponsor personnel were masked to group assignment. To prevent unblinding, CRP values were not provided to the investigators or patients. However, investigators received the CRP PMR-AS value at the end of each patient visit.

The patients received intravenous infusions of tocilizumab (8 mg/kg) or placebo every 4 weeks for 24 weeks (at weeks 0, 4, 8, 12, 16, and 20).

At each study visit after week 8 (weeks 12, 16, and 20), prednisone dosages greater than 10 mg per day were tapered by 5 mg. For dosages less than or equal to 10 mg per day, if the CRP PMR-AS was less than 10, the prednisone dosage was decreased by 2 mg every 2 weeks according to the investigator’s evaluation of disease activity.¹⁶

The primary end point was a composite of low disease activity, defined as CRP PMR-AS less than 10, and either prednisone dosage less than or equal to 5 mg per day or prednisone dosage decrease by greater than or equal to 10 mg vs baseline at 24-week follow-up.

The CRP PMR-AS (range, 0-100; higher scores indicate greater disease activity; no minimal clinically important difference [MCID] defined)^17-20 is based on minutes of morning stiffness multiplied by 0.10, ability to elevate the upper limbs (range, 0-3; higher scores indicate greater ability; no MCID established), physician's global assessment on a 10-point visual analog scale (VAS; 0 indicates best health), patient-assessed pain intensity on a 10-point VAS (0 indicates the least pain), and CRP level in mg/dL. These 5 values were summed to obtain the total score (scores of 7 to 17 indicate moderate disease activity and scores above 17 indicate high disease activity). Treatment deescalation is recommended when the score is less than 10.^16,21-23 For patients with active polymyalgia rheumatica despite prednisone therapy, the objective of a sparing agent is to decrease the dose of prednisone. No guidelines exist regarding the acceptable dosage, but there is a consensus in rheumatology that long-term therapy with less than or equal to 5 mg per day of a prednisone equivalent is acceptable^24,25 and that doses of greater than or equal to 10 mg per day of prednisone induce significant adverse effects.²⁶

There were 11 secondary outcomes assessed at week 24 that are reported in this article: the mean CRP PMR-AS¹⁷; the mean ESR PMR-AS (PMR-AS measured using the ESR instead of the CRP value^13,14,17) (range, 0-100; higher scores indicate worse disease activity; no MCID); the proportion of patients with CRP PMR-AS less than 1.5 (defining remission), less than 10, and less than or equal to 17 (no MCID for any of these outcomes); the glucocorticoid dosage (no MCID defined); the proportion of patients not receiving glucocorticoid therapy at each visit (no MCID defined); the proportion of patients whose prednisone dosage was less than or equal to 5 mg per day at week 24 or at least 10 mg per day lower than at baseline (no MCID defined); and mean scores on the 36-Item Short Form Health Survey (SF-36) mental and physical components (range, 0-100; higher scores indicate better mental and physical condition; MCID, ≥5 points; no specific MCID for polymyalgia rheumatica)²⁷ and Health Assessment Questionnaire Disability Index (range, 0-3; higher scores indicate worse health; no MCID for polymyalgia rheumatica).

Post hoc outcomes (eTable 1 in Supplement 3) assessed at week 12, 16, 20, and 24 included the clinical PMR-AS (range, 0-70; higher scores indicate worse disease activity; no MCID established), calculated as the sum of the 4 clinical items in the PMR-AS.^13,14 The PMR-AS with imputation of the CRP (CRP-imputed PMR-AS) (range, 0-100; higher scores indicate worse disease activity; no MCID established) was assessed at week 24. The percentage of patients with CRP PMR-AS less than 7,¹⁷ the physician’s VAS score for global disease activity (range, 0-10; higher scores indicate worse disease activity; no MCID for polymyalgia rheumatica), and the cumulative prednisone dosage were assessed at each visit. The patient-assessed VAS scores for pain (range, 0-10; higher scores indicate worse pain; MCID for pain, 3 points),²⁸ fatigue, and global disease activity (range, 0-10; higher scores indicate worse disease activity; no MCID established) were collected at each visit.

Other secondary outcomes not reported in this article include ultrasonography scoring of synovitis and tenosynovitis at the hips and shoulders, biological markers of IL-6, TGF-β, immune system cells, CRP, and carboxyterminal region of collagen type I marker for osteoresorption, bone density at the lumbar spine and hip, and the cost/efficacy analysis.

To estimate the required sample size, based on the results of a previous open-label study of tocilizumab in patients with early polymyalgia rheumatica,¹⁰ we selected a 35% between-group difference for the primary end point (ie, achievement of the primary outcome in 25% of the placebo group vs 60% of the tocilizumab group). An improvement of 20% to 40% between groups is considered clinically important in other rheumatic diseases treated by biologics, such as rheumatoid arthritis²⁹ or a recent therapeutical trial in polymyalgia rheumatica.⁶ With the 2-sided α risk set at .05, a total of 90 patients (45 in each group) provided 90% power for detecting a 35% difference in the primary end point (ie, a relative risk of 1.4). Assuming that about 10% of patients would not complete the trial, a sample size of 100 participants (50 in each group) was planned.

The efficacy analyses included all randomized patients who received at least 1 dose of tocilizumab or placebo. The primary analysis compared the percentage of patients who met criteria for the composite primary outcome in the 2 groups at the 24-week follow-up. Patients with missing data due to premature trial discontinuation for lack of efficacy were classified as not having met the primary end point according to a blinded review of the patient’s data by the scientific committee. A sensitivity analysis was planned to impute the remaining missing data as not meeting the primary outcome. Missing data for other outcomes were disregarded.

The proportion of patients with the primary end point was compared between the 2 groups using a generalized linear model with binomial distribution, logarithmic link (producing the adjusted relative risk [aRR]), and exchangeable correlation structure to take a potential center effect into account. Adjusted (for center) differences were obtained using an identity link. The same method was applied for all secondary outcomes, using a normal distribution and an identity link for continuous outcomes (producing adjusted mean differences).

Analyses were 2-sided with α of .05. Because of the potential for type I error due to multiple comparisons, findings for analyses of secondary end points should be interpreted as exploratory. SAS, version 9.4 (SAS Institute), was used for all analyses.

Patients were enrolled from February 2017 to October 2019. Final follow-up occurred on November 12, 2020. Of 113 patients screened for eligibility, 101 were randomized for treatment. One patient was excluded after randomization before receiving the study drug because he developed giant cell arteritis. Overall, 100 patients received at least 1 dose of tocilizumab (49 patients) or placebo (51 patients) and were included in the analyses (Figure 1). Baseline characteristics of randomized patients are shown in Table 1.

At the 24-week follow-up, the composite primary end point occurred in 67.3% of patients treated with tocilizumab and 31.4% treated with placebo (adjusted difference, 36.0% [95% CI, 19.4%-52.6%]; aRR, 2.3 [95% CI, 1.5-3.6]; P < .001) (Table 2). Two patients in the tocilizumab group missed the 24-week follow-up and were classified as “treatment failures” during blinded review. Therefore, there were no missing data and a sensitivity analysis regarding missing data was not performed.

At the 24-week follow-up, mean CRP PMR-AS values were 7.5 in the tocilizumab group and 14.9 in the placebo group (adjusted mean difference, −7.5 [95% CI, −11.2 to −3.8]) and mean ESR PMR-AS values were 8.1 in the tocilizumab group and 15.9 in the placebo group (adjusted mean difference, −7.8 [95% CI, −12.1 to −3.4]) (Table 2). The percentage of patients with remission (CRP PMR-AS ≤1.5) was not significantly different between the groups (7 patients [14.3%] vs 4 patients [7.8%]; adjusted mean difference, 6.9% [95% CI −0.02% to 0.16%]; aRR, 1.8 [95% CI, 0.8-3.9]).

A significantly greater percentage of patients in the tocilizumab group, compared with the placebo group, had a CRP PMR-AS less than 10 (36 patients [73.5%] vs 22 patients [43.1%]; adjusted difference, 31.1% [95% CI, 12.2%-50.0%]; aRR, 1.8 [95% CI, 1.22.7]) and less than or equal to 17 (42 [89.4%] vs 33 [64.7%]; adjusted difference. 26.0% [95% CI, 13.3%-39.0%; aRR, 1.4 [95% CI, 1.2-1.7]) (Table 2).The mean prednisone dose at week 24 was significantly lower with tocilizumab compared with placebo (3.8 mg/d vs 6.1 mg/d; adjusted mean difference, −2.3 [95% CI, −3.6 to −1.0] mg/d; P < .001). At week 24, the percentage of patients no longer receiving glucocorticoid therapy (24 patients [49%] vs 10 patients [19.6%]; adjusted difference 29.3% [95% CI, 18.9%-39.7%]; aRR, 2.5 [95% CI, 1.8-3.5]) and the percentage of patients with a glucocorticoid dosage less than 5 mg per day or a decrease of at least 10 mg per day vs baseline (37 patients [75.5%] vs 26 patients [51.1%]; adjusted difference, 25.7% [95% CI, 12.1%-39.3%]; aRR, 1.6 [95% CI, 1.2-2.1]) were each significantly higher in the tocilizumab group compared with the placebo group (Table 2). Scores were not significantly better with tocilizumab than with placebo for the physical (37.9 vs 36.4; adjusted mean difference, 1.6 [95% CI, −2.4 to 5.7 ]; P = .43) and mental (42.2 vs 41.0; adjusted mean difference, 1.2 [95% CI, −3.1 to 5.5 ]; P = .59) components of the SF-36 or the Health Assessment Questionnaire Disability Index (1.6 vs 1.7; adjusted mean difference, −0.1 [95% CI, −0.3 to 0.1]; P = .32).

Mean clinical PMR-AS and mean CRP-imputed PMR-AS were significantly lower in the tocilizumab group compared with the placebo group. Values for clinical PMR-AS were 7.4 vs 13.5 (adjusted mean difference, −6.1; [95% CI, −10.1 to −2.1]; P = .003) and values for mean CRP-imputed PMR-AS were 8.6 vs 15.4 (adjusted mean difference, −6.8 [95% CI, −11.3 to −2.4]; P = .003). The percentage of participants who had CRP PMR-AS less than 7 was significantly greater in the tocilizumab group compared with the placebo group (29 [59.2%] vs 17 [33.3%]; adjusted difference, 25.8% [95% CI, 7.4%-44.3%]; aRR, 1.8 [95% CI, 1.1-2.9]) (Figure 2; eTable 1 in Supplement 3). The mean physician VAS score for global disease activity was significantly lower in the tocilizumab group compared with the placebo group (1.6 vs 3.1; adjusted mean difference, −1.5 [95% CI, −2.5 to −0.5]; P = .004). The mean cumulative prednisone dosage was significantly lower in the tocilizumab group than in the placebo group (1380 mg vs 1591 mg; adjusted difference, −207 [95% CI, −343 to −71] mg; P = .003). The patient-reported VAS scores for pain, fatigue, and global disease activity were not significantly different between the 2 groups at week 24 (eTable 1 in Supplement 3).

Infections were the most common adverse events, occurring in 23 patients (46.9%) in the tocilizumab group and 20 (39.2%) in the placebo group (Table 3). Headache occurred in 7 patients (14.3%) in the tocilizumab group and in 1 patient (2.0%) in the placebo group (Table 3).

Herpes zoster occurred in 1 patient in each group and resolved without discontinuation of the assigned treatment. There were no deaths. Serious adverse events are reported in eTable 2 and eTable 3 in Supplement 3.

In this clinical trial of patients with glucocorticoid-dependent polymyalgia rheumatica, tocilizumab, compared with placebo, resulted in a greater proportion of patients meeting the primary outcome of a CRP PMR-AS less than 10 combined with either a prednisone dosage less than or equal to 5 mg per day or a decrease greater than or equal to 10 mg from baseline at the 24-week follow-up.

Previous studies have reported that tocilizumab was associated with improvement in polymyalgia rheumatica disease activity in patients with recent-onset polymyalgia rheumatica.^{10,11,13,31-36} In contrast to these studies, the current randomized clinical trial included only patients dependent on glucocorticoid therapy. In this trial, glucocorticoid dependency was defined as a flare when the prednisone dosage fell below 10 mg per day during a taper and is consistent with the definition of glucocorticoid dependency as defined by an international committee. Approximately half of patients with polymyalgia rheumatica have relapses after glucocorticoid discontinuation and take prednisone in dosages of 10 mg per day or more.^3,4

The primary outcome was a composite outcome that included both low disease activity and lower prednisone requirements compared with baseline. The CRP PMR-AS measure of disease activity is the only currently used and validated composite disease activity measure to assess polymyalgia rheumatica disease activity.^6,10,12 This score is simple to use and valid when measured by rheumatologists or general practitioners, although there is interobserver variability in physician-assessed global disease activity.^21-23 This score was used for tailoring the glucocorticoid dosage to disease activity.²³

Disease activity decreased in the placebo group during the first few weeks after randomization, likely because prednisone dose tapering was not attempted until after week 8 of the trial. Consistent with this aspect of the study design, the effect of tocilizumab on the prednisone dosage became apparent only after the first 8 weeks. Although physician assessments documented improved outcomes in the tocilizumab group compared with the placebo group, patient-reported outcomes were not significantly better in the tocilizumab group compared with the placebo group. This discrepancy suggests that patients with polymyalgia rheumatica who are dependent on glucocorticoids may not be aware of improvement from tocilizumab and requires further study. It also suggests that benefits from tocilizumab may not be large.

Rates of adverse events were similar between the groups. However, the trial did not have statistical power to detect significant differences in adverse events between the 2 groups.

This study has several limitations. First, the exclusion criteria included many health conditions and laboratory test abnormalities. Second, only patients with glucocorticoid-dependent polymyalgia rheumatica were included. Third, prednisone toxicity was not measured using a validated tool. The only available tool is the Glucocorticoid Toxicity Index, which was published in March 2017, shortly after the start of patient enrollment.³⁷ Fourth, CRP PMR-AS greater than 10 may be viewed as an inappropriate inclusion criterion because a decrease by only 2 points would result in a patient meeting a component of the primary outcome (CRP PMR-AS <10). However, the primary outcome also required prednisone sparing, and even a small CRP PMR-AS decrease likely constitutes a meaningful improvement if combined with a prednisone-sparing effect.^6,11,25 Fifth, the Chuang Classification criteria¹⁵ were used for inclusion in the SEMAPHORE study even though provisional criteria of European League Against Rheumatism/American College of Rheumatology were suggested in 2012.³⁰ Sixth, CRP was among the 5 items of the CRP PMR-AS included in the composite primary end point, which may have induced evaluation bias because a decrease of CRP could lower the PMR-AS. However, the contribution of CRP to the overall score is small, with a 1-mg/dL decrease resulting in a score decrease of only about 1 point on the scale ranging from 0 to 100. Seventh, disease duration was longer and methotrexate therapy was more common in those randomized to receive tocilizumab. Eighth, the length of follow-up for the primary end point was relatively short (week 24). Ninth, the primary outcome has not been fully validated.

Among patients with active polymyalgia rheumatica despite prednisone therapy, tocilizumab, compared with placebo, resulted in a significantly greater proportion of patients with a CRP PMR-AS less than 10 with reduced prednisone requirements at week 24. Further research is needed to confirm efficacy and to determine the balance of potential benefits and harms.

Accepted for Publication: August 18, 2022.

Corresponding Author: Valerie Devauchelle-Pensec, MD, PhD, Rheumatology Unit, Hôpital de la Cavale Blanche, BP 824, F 29609 Brest Cedex, France (valerie.devauchelle-pensec@chu-brest.fr).

Author Contributions: Drs Devauchelle-Pensec and Saraux had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Devauchelle-Pensec, Hilliquin, Nowak, Saraux.

Acquisition, analysis, or interpretation of data: Devauchelle-Pensec, Carvajal Alegria, Dernis, Richez, Truchetet, Wendling, Toussirot, Perdriger, Gottenberg, Felten, Fautrel, Chiche, Le Henaff, Dervieux, Direz, Chary-Valckenaere, Cornec, Guellec, Marhadour, Nowak, Saraux.

Drafting of the manuscript: Devauchelle-Pensec, Carvajal Alegria, Hilliquin, Nowak, Saraux.

Critical revision of the manuscript for important intellectual content: Devauchelle-Pensec, Dernis, Richez, Truchetet, Wendling, Toussirot, Perdriger, Gottenberg, Felten, Fautrel, Chiche, Le Henaff, Dervieux, Direz, Chary-Valckenaere, Cornec, Guellec, Marhadour, Nowak, Saraux.

Statistical analysis: Devauchelle-Pensec, Nowak, Saraux.

Obtained funding: Devauchelle-Pensec, Le Henaff, Saraux.

Administrative, technical, or material support: Devauchelle-Pensec, Carvajal Alegria, Truchetet, Guellec, Saraux.

Supervision: Devauchelle-Pensec, Richez, Wendling, Gottenberg, Hilliquin, Saraux.

Other - selection & interpretation: Richez.

Other - clinical and ultrasonographic investigator: Chary-Valckenaere.

Conflict of Interest Disclosures: Dr Devauchelle-Pensec reported receiving personal fees from Chugai Pharmaceutical and AbbVie and grants and personal fees from Novartis during the conduct of the study and personal fees from Galapagos, Pfizer, Bristol Myers Squibb, Janssen, and AbbVie outside the submitted work. Dr Carvajal Alegria reported receiving personal fees from Chugai Pharmaceutical outside the submitted work. Dr Dernis reported receiving personal fees from Novartis, Bristol Myers Squibb, UCB, AbbVie, Nordic Pharma, Amgen, and Janssen outside the submitted work. Dr Richez reported receiving personal fees from Sanofi, personal fees from Lilly, and personal fees from Novartis outside the submitted work. Dr Wendling reported personal fees from AbbVie, Bristol Myers Squibb, MSD, Pfizer, Chugai Pharmaceutical, Amgen, Nordic Pharma, UCB, Novartis, Janssen, Eli Lilly, Grunenthal, and Galapagos outside the submitted work. Dr Gottenberg reported receiving personal fees from Roche-Chugai Pharmaceutical, Sanofi, Pfizer, Eli Lilly, Gilead, and Abbvie and grants from Bristol Myers Squibb outside the submitted work. Dr Fautrel reported receiving personal fees from Roche Chugai Pharmaceutical during the conduct of the study. Dr Direz reported receiving personal fees from Novartis and personal fees from Roche-Chugai Pharmaceutical outside the submitted work. Dr Nowak reported receiving grants from the French National Program for Clinical Research and nonfinancial support from Roche-Chugai Pharmaceutical, which donated the infusion form of tocilizumab during the conduct of the study. Dr Saraux reported receiving grants from Roche-Chugai Pharmaceutical and grants from the French National Program for Clinical Research during the conduct of the study and personal fees from Nordic Galapagos, AbbVie, Eli Lilly, Nordic, and Roche-Chugai and grants from Eli Lilly outside the submitted work. No other disclosures were reported.

Funding/Support: The funders of this study were the French National Program for Clinical Research and the study sponsor (ie, the Brest University Hospital). Roche-Chugai Pharmaceutical provided an unconditional grant for the study and donated the tocilizumab used for the study.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. They had no right to veto publication or to influence the choice of the journal to which the paper was submitted.

Data Sharing Statement: See Supplement 4.

Additional Contributions: We thank the French rheumatologists, particularly those from the French University Hospital VICTOR HUGO (InnoVation en reCherche OsTeo-aRticulaire des Hôpitaux Universitaires du Grand Ouest) network (srouest.fr), and the general practitioners who referred their patients to this trial. We thank Mr Sidney Marquez, professional medical writer, who received compensation for assisting with writing the manuscript. We are grateful to the Clinical Investigations Center (CIC) 1412, Institut National de la Santé et de la Recherche Médicale (INSERM), Brest, France, for centralizing the trial data and to Audrey le Goff, MS; Adrien Clarysse, MS; and Valentine Guiton, MS, of the Brest University Hospital research board (DRCI) at the Brest University Hospital, who received no compensation for their role in the study.