This is a prospective, multicenter, open-label randomized controlled trial (RCT) designed to evaluate and generate evidence on the optimal combination of TGTs with other csDMARDs (methotrexate, leflunomide, and hydroxychloroquine). The trial is being conducted at 3 sites in China: Dongzhimen Hospital, Beijing University of Chinese Medicine; Acupuncture and Moxibustion Hospital, China Academy of Chinese Medical Sciences; and Wuhan Hospital of Traditional Chinese Medicine. Eligible patients with RA were enrolled and assigned via block randomization using SAS (version 9.3; SAS Institute). Participants were allocated in a 1:1:1:1 ratio to 1 of 4 treatment groups: group 1 (TGT monotherapy), group 2 (TGTs+methotrexate), group 3 (TGTs+hydroxychloroquine), and group 4 (TGTs+leflunomide). The total observation period was 12 weeks, including a 4-week treatment period, an 8-week treatment period, and a 12-week treatment period. The flowchart of this study is shown in Figure 1, and the evaluation time points are shown in Textbox 1.

Patients who met the 1987 American College of Rheumatology (ACR) [20] or the 2010 ACR/European League Against Rheumatism [21] classification criteria for RA were eligible for inclusion. The detailed inclusion and exclusion criteria are provided in Textbox 1.

Participants were recruited primarily from the Department of Rheumatology at Dongzhimen Hospital of Beijing University of Chinese Medicine, the Department of Rheumatology at Acupuncture and Moxibustion Hospital at China Academy of Chinese Medical Sciences, and the Department of Rheumatology at Wuhan Hospital of Traditional Chinese Medicine, where research assistants assisted in screening participants. Additional participants were recruited through advertisements such as job postings.

Eligible patients with RA were enrolled and assigned to 4 treatment groups using block randomization with a 1:1:1:1 allocation ratio. Statistical experts who were not involved in the study or data analysis generated the block randomization sequence using the SAS software.

This is an open-label study; therefore, neither researchers nor participants were blinded. Only adjudicators of outcome measures and statistical analysts were blinded to group allocation.

All participants received targeted TGT-based therapy, administered either as monotherapy or in combination with methotrexate (10‐15 mg/wk), leflunomide (10‐20 mg/d), or hydroxychloroquine (200‐400 mg/d). Participants were evaluated at baseline and then every 4 weeks for a total of 12 weeks.

For ethical reasons, nonsteroidal anti-inflammatory drugs (NSAIDs), such as diclofenac sodium enteric-coated tablets, were permitted as rescue therapy when pain became intolerable during the study. The permitted regimen was 25 mg orally 3 times daily before meals, and the use, dosage, and therapeutic effects were recorded in detail.

However, during the trial period, all other medications with similar mechanisms or effects to the study drug were prohibited, except for rescue therapy for intolerable RA pain. This included glucocorticoids (eg, prednisone acetate and methylprednisolone), biologic agents (eg, adalimumab and secukinumab), and other disease-modifying antirheumatic drugs (eg, cyclophosphamide and cyclosporine).

Safety will be evaluated through routine blood tests, as well as renal and hepatic function tests.

Drug-related adverse events, including infections, liver dysfunction, kidney dysfunction, and treatment-related complications (eg, injection site reactions), were recorded. Serious adverse events were documented separately. Adverse events will be summarized by frequency (eg, number and percentage of participants affected) and severity according to the CTCAE criteria.

Patients’ adherence was assessed by counting the number of remaining cartridges at the end of treatment. In addition, dropout status and reasons for discontinuation were recorded throughout the 12-week observation period.

Patients and the public were not involved in the development of the research question or study design. They will not participate in trial recruitment, conduct, or reporting of results. Trial results will be disseminated to participants via social media.

The sample size was calculated for a multicenter RCT with 1 common control group (TGT monotherapy) and 3 active treatment groups (TGTs+methotrexate, TGTs+hydroxychloroquine, and TGTs+leflunomide). The primary hypothesis is that each combination therapy will be superior to TGTs monotherapy for the ACR20 response rate at week 12. The assumed ACR20 response rates were derived from an unpublished study conducted at our center in 2022 (Liu Lv, unpublished data, May 2022). This study included 30 patients with moderate RA (DAS28 3.2-5.1), randomized 1:1 to receive either TGT monotherapy (n=15) or TGTs+methotrexate (n=15) for 12 weeks. The observed ACR20 response rates were 53.3% (8/15) in the monotherapy group and 80% (12/15) in the combination group. On the basis of these data, the following assumptions were made:

We consider this absolute difference of 26 percentage points to be clinically meaningful.

The sample size calculation was performed using PASS (Power Analysis and Sample Size) software (version 15; NCSS, LLC). The procedure selected was “chi-square tests” to compare multiple treatment groups with a common control while controlling the family-wise type I error rate.

Input parameters included the following: solving for sample size, degree of freedom 3; power 0.8; α=.05, and effect size (w=0.255098).

Consequently, the Dunnett procedure required 42 participants per group, resulting in a total sample size of 168 before accounting for dropouts. With an estimated dropout rate of 10%, the adjusted total sample size was approximately 188 participants, corresponding to approximately 47 participants per group.

Outcome assessors at each center recorded all study data. These assessors received standardized training prior to participant recruitment. The training covered procedures for accurately and promptly completing case report forms (CRFs) and electronic CRFs. Standard operating procedures were established and made available to investigators and assessors at each trial site.

Investigators were not permitted to modify or access data until enrollment, follow-up, and data collection had been completed for all participants.

The primary outcome—the ACR20 response rate at week 12—will be compared between each combination therapy group and the TGT monotherapy group using logistic regression. The study center will be included as a fixed effect to account for potential site differences.

Secondary outcomes (eg, DAS28-ESR and CDAI) will be analyzed using mixed-effects models for repeated measures. These models will include fixed effects for treatment, visit, treatment-by-visit interaction, baseline value, and study center, with participants included as a random intercept.

All primary analyses will be conducted on the intention-to-treat population, defined as all randomized participants who received at least 1 dose of the study drug. Missing data will be handled using multiple imputation under the missing-at-random assumption, generating 20 imputed datasets. Sensitivity analyses will be performed to evaluate the robustness of conclusions under different missing data assumptions.

Safety data, including adverse events, will be summarized descriptively by treatment group. Between-group comparisons of adverse event rates will be conducted using the Fisher exact test.

Rescue therapy with NSAIDs was permitted for intolerable joint pain, in line with ethical requirements and standard clinical practice for the management of RA. As pain and swelling are universal symptoms of active RA, the requirement for rescue analgesia reflects real-world treatment conditions and will not be statistically adjusted in the primary analysis. The use of rescue medication will be documented descriptively and reported by treatment group. If notable between-group differences in the frequency of rescue therapy are observed, exploratory sensitivity analyses will be conducted to assess their potential influence on efficacy outcomes.

All analyses will be performed using SAS (version 9.4) and R (version 4.4.s2; R Foundation for Statistical Computing). Between-group differences will be reported with 95% CIs. Statistical significance is set at α=.05 (2-sided).

All investigators received specific training on trial objectives, protocols, treatment strategies, and quality control procedures before the trial commenced. All investigators involved in this study were rheumatologists with at least 5 years of clinical experience.

All trial documents, including screening forms, CRFs, and treatment records, are stored at the study site under locked conditions with restricted access.

Every 3 months, members of the quality monitoring team will conduct on-site quality control reviews at each participating site and generate reports on the overall quality of the study processes. Trial managers will hold regular meetings to discuss and address any issues identified during monitoring.

The Dongzhimen Hospital Ethics Committee of Beijing University of Chinese Medicine approved the study protocol (2024DZMEC-241). The trial is conducted in accordance with the Declaration of Helsinki and is registered with the Chinese Clinical Trial Registry.

Each participant was informed of the study’s purpose, potential risks, and benefits. The recruiting physician reviewed the written informed consent form with the participant. Participants did not receive any financial compensation for their involvement in this study, as the trial involved only marketed drugs with well-established safety and efficacy profiles, and no additional procedures beyond routine clinical practice were required.

In this study, personal identifying information was removed and replaced with a unique, nonidentifiable code. No personal information was transmitted, ensuring data confidentiality. The results of this study will be presented at selected conferences and scientific meetings and published in peer-reviewed journals. To promote transparency, the original dataset will be shared with external researchers upon reasonable request, beginning three years after the main results are published.

This trial is expected to provide robust evidence on the comparative efficacy and safety of TGT-based combination therapies in patients with moderately active RA. We hypothesize that all 3 TGT combination regimens (TGTs+methotrexate, TGTs+hydroxychloroquine, and TGTs+leflunomide) will result in significantly higher ACR20 response rates at week 12 than TGT monotherapy. These findings are expected to clarify the risk-benefit profile of these combinations and identify which regimen offers the most favorable balance of efficacy and tolerability for different patient subgroups.

The trial focuses on individuals with DAS28 scores between 3.2 and 5.1. This subgroup is targeted because early intervention with csDMARD combinations may prevent disease progression and reduce long-term disability. TGTs, a cornerstone of traditional Chinese medicine, have demonstrated efficacy comparable to methotrexate in monotherapy. However, their role in combination regimens remains poorly established [15,17,25]. This trial directly compares TGTs with 3 commonly used csDMARDs (methotrexate, leflunomide, and hydroxychloroquine), providing insights into synergistic effects, safety trade-offs, and optimal dosing strategies. The 12-week intervention period is designed to detect early treatment responses, which are critical for patients requiring rapid symptom control and for clinicians making timely therapeutic adjustments. Additionally, excluding participants with reproductive plans ensures a focus on the safety profile of TGTs in populations less vulnerable to their known gonadal toxicity, thereby refining the risk-benefit assessment of TGTs in clinical practice.

The results of this trial could reshape RA treatment paradigms, especially in settings with limited resources where biologic agents are unavailable. If TGT-based combination therapy is shown to be noninferior or superior to monotherapy, it could become a cost-effective initial treatment strategy, reducing the need for expensive biologics. Furthermore, the results of this study will provide evidence-based support for personalized treatment decisions involving TGT combination regimens.

Exclusion of patients with reproductive potential may discourage recruitment, resulting in an eligible population composed predominantly of middle-aged and older individuals. To mitigate this issue, the trial was supported by a social media campaign that highlighted the study’s focus on safety in the nonreproductive population.

Compliance is another concern, as the risk of gastrointestinal side effects and hepatotoxicity associated with TGTs may lead to early discontinuation. To address this, active measures were implemented, including standardized patient education, routine monitoring of liver enzymes, protocol-based dose adjustments, aggressive follow-up, intensive training of participating investigators, and the use of digital joint counting tools to standardize scores.

Finally, because neither the investigators nor participants were blinded, and only outcome assessors and statistical analysts were blinded to group allocation, there is a potential risk of bias in the results.

This prospective, multicenter, open-label RCT will provide robust evidence on the efficacy and safety of TGTs in combination with csDMARDs for the treatment of RA with moderate disease activity.