Stratification of biological therapies by pathobiology in biologic-naive patients with rheumatoid arthritis (STRAP and STRAP-EU): two parallel, open-label, biopsy-driven, randomised trials
Bibliographic record
Abstract
BACKGROUND: Despite highly effective targeted therapies for rheumatoid arthritis, about 40% of patients respond poorly, and predictive biomarkers for treatment choices are lacking. We did a biopsy-driven trial to compare the response to rituximab, etanercept, and tocilizumab in biologic-naive patients with rheumatoid arthritis stratified for synovial B cell status. METHODS: STRAP and STRAP-EU were two parallel, open-label, biopsy-driven, stratified, randomised, phase 3 trials done across 26 university centres in the UK and Europe. Biologic-naive patients aged 18 years or older with rheumatoid arthritis based on American College of Rheumatology (ACR)-European League Against Rheumatism classification criteria and an inadequate response to conventional synthetic disease-modifying antirheumatic drugs (DMARDs) were included. Following ultrasound-guided synovial biopsy, patients were classified as B cell poor or B cell rich according to synovial B cell signatures and randomly assigned (1:1:1) to intravenous rituximab (1000 mg at week 0 and week 2), subcutaneous tocilizumab (162 mg per week), or subcutaneous etanercept (50 mg per week). The primary outcome was the 16-week ACR20 response in the B cell-poor, intention-to-treat population (defined as all randomly assigned patients), with data pooled from the two trials, comparing etanercept and tocilizumab (grouped) versus rituximab. Safety was assessed in all patients who received at least one dose of study drug. These trials are registered with the EU Clinical Trials Register, 2014-003529-16 (STRAP) and 2017-004079-30 (STRAP-EU). FINDINGS: Between June 8, 2015, and July 4, 2019, 226 patients were randomly assigned to etanercept (n=73), tocilizumab (n=74), and rituximab (n=79). Three patients (one in each group) were excluded after randomisation because they received parenteral steroids in the 4 weeks before recruitment. 168 (75%) of 223 patients in the intention-to-treat population were women and 170 (76%) were White. In the B cell-poor population, ACR20 response at 16 weeks (primary endpoint) showed no significant differences between etanercept and tocilizumab grouped together and rituximab (46 [60%] of 77 patients vs 26 [59%] of 44; odds ratio 1·02 [95% CI 0·47-2·17], p=0·97). No differences were observed for adverse events, including serious adverse events, which occurred in six (6%) of 102 patients in the rituximab group, nine (6%) of 108 patients in the etanercept group, and three (4%) of 73 patients in the tocilizumab group (p=0·53). INTERPRETATION: In this biologic-naive population of patients with rheumatoid arthrtitis, the dichotomic classification into synovial B cell poor versus rich did not predict treatment response to B cell depletion with rituximab compared with alternative treatment strategies. However, the lack of response to rituximab in patients with a pauci-immune pathotype and the higher risk of structural damage progression in B cell-rich patients treated with rituximab warrant further investigations into the ability of synovial tissue analyses to inform disease pathogenesis and treatment response. FUNDING: UK Medical Research Council and Versus Arthritis.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.012 | 0.010 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.005 | 0.005 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.004 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".