Researchers have yet to crack how to analyze synovial tissue biopsy specimens to predict treatment responses for patients with rheumatoid arthritis.
According to new research, classifying patients by synovial fluid B-cell status was not reliable in predicting treatment response to etanercept, tocilizumab, or rituximab. More comprehensive molecular analyses may hold promise in developing models to predict treatment of a disease as heterogeneous as RA, wrote the authors, led by chief investigator Costantino Pitzalis, MD, head of the Centre for Experimental Medicine and Rheumatology at Queen Mary University of London.
The article was published in the November 2023 issue of The Lancet Rheumatology..
Trial-and-error treatment
Because clinicians do not have a way to reliably predict how patients may respond to certain medications, the current RA treatments involve trial and error.
Courtesy Dr. Arthur Mandelin, Northwestern University
A synovial tissue biopsy procedure is shown in progress.
“Both with regard to the first conventional synthetic disease-modifying antirheumatic drug (DMARD) for all patients and the first biologic DMARD for the subgroup of patients requiring these drugs, the choice of a drug is based on pragmatic arguments rather than on individual patient characteristics,” Annette H. van der Helm-van Mil, MD, PhD, professor of rheumatology at the Leiden University Medical Center, the Netherlands, wrote in an accompanying editorial.
Most research on predicting treatment response has used clinical features and blood biomarkers, but neither approach is reliably predictive. “Precision medicine, therefore, remains elusive,” she said. Newer research has focused on synovial tissue biopsy to inform research decisions.
The STRAP studies
In this newest study, researchers combined data from two clinical studies with identical design: the Stratification of Biological Therapies by Pathobiology in Biologic-Naive Patients With Rheumatoid Arthritis (STRAP) trial, taking place in the United Kingdom, and STRAP-EU, which recruited patients from the European Union. The trials are the largest biopsy-driven trials to date, according to the researchers.
In total, researchers recruited 223 biologic-naive adult patients from 26 universities in the United Kingdom and Europe. Participants underwent ultrasound-guided synovial tissue biopsy and were then randomly assigned to receive etanercept (72 patients), tocilizumab (73 patients), or rituximab (78 patients). In a histologic analysis, 121 patients were characterized as B-cell poor (BCP), 100 patients were classified as B-cell rich (BCR), and two patients were classified as undetermined.
“We hypothesized that patients with a low synovial histological or molecular B-cell score would have a lower response to rituximab than to etanercept or tocilizumab,” the authors wrote.
However, among the BCP patients, there were no significant differences between responses to treatment at week 16 in the rituximab group compared with the etanercept and tocilizumab groups put together. In both groups, around 60% of patients achieved the primary endpoint of at least 20% improvement in American College of Rheumatology response criteria.
The results suggest that “a dichotomic classification into synovial B cell poor versus rich is unable to predict treatment response in patients treated with rituximab compared with etanercept or tocilizumab,” the authors wrote.
The findings echo work from the same group of researchers in 2021. This study – the R4RA trial – enrolled 164 patients who previously had an inadequate response to tumor necrosis factor blockers. Researchers then used a similar histologic approach to compare patients’ responses to tocilizumab or rituximab.
Among patients classified as BCP, there was not a significant difference in treatment response between the tocilizumab group and the rituximab group. However, classifying patients as BCP with RNA sequencing did make a difference. In this subgroup, patients given tocilizumab demonstrated a significantly higher response rate than those treated with rituximab.