Data Sources

This post-hoc analysis was conducted using data from a randomized, multicenter, double-blind, parallel-group, placebo-controlled phase II/III trial of OZR with MTX in Japanese patients with RA (jRCT2080223971, OHZORA trial) and a randomized, multicenter, open-label, phase III trial without MTX (jRCT2080223973, NATSUZORA trial). Details of the study designs, patient populations, and results of these trials have been published previously [14,15,16]. In the OHZORA trial, 381 patients were assigned to receive placebo or 30 mg or 80 mg of OZR in a 1:2:2 ratio. The treatments were subcutaneously administered with MTX every 4 weeks for 24 weeks (period A) [14]. After 24 weeks in this study, all patients received either 30 mg or 80 mg of OZR (period B) [14, 15]. In the NATSUZORA trial, OZR (30 mg:80 mg in a 2:1 ratio) was administered subcutaneously every 4 weeks for 52 weeks without MTX [16].

In the full analysis set of the two studies, patients who met the following criteria were included in the current post-hoc analyses:

1.

Placebo group in the OHZORA trial (continued MTX monotherapy; “MTX group”)

2.

OZR (30 mg) group in the OHZORA trial (with MTX; “OZR + MTX group”)

3.

OZR (30 mg) group in the NATSUZORA trial (without MTX; “OZR group”).

Two PSM, defined below as PSM-A and PSM-B, were performed and analyzed. The target patient groups for each analysis are indicated.

PSM-A::

75 patients in the MTX group of the OHZORA trial (with MTX) and 94 patients in the OZR group of the NATSUZORA trial (without MTX)

PSM-B::

152 patients in the OZR+MTX group of the OHZORA trial (with MTX) and 94 patients in the OZR group of the NATSUZORA trial (without MTX).

The scheme is shown in Fig. 1.

Fig. 1

Design of each clinical trial of OZR and matching analysis. Two types of matching analyses were performed using three groups from two clinical trials: PSM-A is an indirect comparison between the OZR and MTX groups; PSM-B is an indirect comparison between the OZR and OZR + MTX groups. csDMARDs conventional synthetic disease-modifying antirheumatic drugs, EE early escape, IR inadequate response, NA not applicable, MTX methotrexate, OZR ozoralizumab, PBO placebo, PSM propensity score matching, RA rheumatoid arthritis. *OZR was administered subcutaneously once every 4 weeks. **Patients who continued MTX treatment for at least 12 weeks

Efficacy Outcomes

Measurements and changes from baseline were calculated for each evaluation period (time point) for the following endpoints: American College of Rheumatology (ACR) 20/50/70 improvement rate (by nonresponder imputation, NRI), patient’s global assessment using a visual analog scale (VAS), physician’s global assessment using a VAS, patient’s pain assessment using a VAS (pain VAS), tender joint count (TJC), swollen joint count (SJC), high-sensitivity C-reactive protein (hs-CRP), erythrocyte sedimentation rate (ESR), clinical disease activity index (CDAI), simplified disease activity index (SDAI), and health assessment questionnaire disability index (HAQ-DI).

In PSM-A, the MTX group in the OHZORA trial (with MTX) and the OZR group in the NATSUZORA trial were included up to 24 weeks because the blinding period was 24 weeks. In PSM-B, the data were tabulated from the start of treatment to 52 weeks. In the OHZORA trial, early escape (EE) criteria were established to rescue patients with an inadequate response, and those patients were placed in period B after 20 weeks of efficacy evaluation. Therefore, data after 24 weeks were excluded from the analysis for cases with EE, and data up to 20 weeks were used in the analysis.

The longitudinal data analysis of changes from baseline compared between treatment groups was conducted using an MMRM based on the assumption of missing at random.

Statistical Analysis

Data were analyzed using SAS version 9.4 (SAS Institute, Tokyo, Japan). The PSMATCH procedure in SAS was used to determine eligible cases. The covariate selection process included age, sex, and RA duration as essential variables, while other covariates were selected based on their clinical importance, including in real-world clinical practice. Considering that body weight was selected as a covariate in the population pharmacokinetic analysis of OZR [17], weight was used in this study instead of body mass index. Finally, the baseline values of the following variables were used as covariates: age, sex, duration of disease, TJC (28 joints), SJC (28 joints), physician’s global assessment using a VAS, patient’s global assessment using a VAS, hs-CRP, weight, usage of glucocorticoids, and CDAI. The method of the match statement in SAS was set to Greedy (K = 1), and Caliper was set to 0.20.

For patient background values (items that were covariates in PSM), the standardized mean difference (SMD) was used to assess the balance of patient characteristics between groups, with differences < 0.1 considered as balanced.

For continuous variables, the mean and associated standard deviation (SD) or median and associated interquartile range (IQR) were calculated. Student’s t test was used to test the efficacy endpoints, and the Kruskal–Wallis test was used to examine the equality of distribution across groups. For the categorical variables, frequency tabulations were performed, and the chi-square test was used to examine the equality of distribution across groups. The significance level was set at 15% (two-tailed) for tests of equality of the distribution of participant background items and at 5% (two-tailed) for other tests, unless otherwise specified.

The change from baseline at weeks 1–52 was tested using an MMRM with treatment group, period, and their interactions as the fixed effects and baseline as the covariate. The MMRM variance–covariance matrix was set to “unstructured.” However, because the hs-CRP values in PSM-B did not converge, “compound symmetry” was used. The Kenward–Roger method was used to calculate the degrees of freedom.

Compliance with Ethics Guidelines

This study is based on previously conducted studies and does not contain new studies with human participants performed by the authors. All studies were conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonization Good Clinical Practice guidelines. Each study was approved by ethics committees, as described in the original papers [14,15,16]. Written informed consent was obtained from all patients.