NEW YORK – The gene expression and histological features found in synovial biopsy samples may hold clues for predicting treatment responses in rheumatoid arthritis (RA), new research suggests, potentially making it possible to select a drug that is more apt to work in a patient based on the specific features found in their disease.
"Incorporating molecular information prior to prescribing arthritis treatments to patients could forever change the way we treat the condition. Patients would benefit from a personalized approach that has a far greater chance of success, rather than the trial-and-error drug prescription that is currently the norm," co-senior and co-corresponding author Costantino Pitzalis, a rheumatology researcher at Queen Mary University of London, said in a statement.
Pitzalis added that "incorporation of these signatures in future diagnostic tests will be a necessary step to translate these findings into routine clinical care."
For a paper appearing in Nature Medicine on Thursday, members of the R4RA Collaborative Group did RNA sequencing and deep histopathological profiling on 164 RA patients from the R4RA, a randomized clinical trial focused on finding features that coincided with better or worse responses to the B-cell targeting anti-CD20 monoclonal antibody treatment rituximab or the interleukin IL-6 receptor-targeting monoclonal antibody tocilizumab.
"[I]n-depth histological/molecular analyses of R4RA synovial biopsies identify humoral immune response gene signatures associated with response to rituximab and tocilizumab, and a stroma/fibroblast signature in patients refractory to all medications," the authors reported, adding that each drug seemed to have distinct post-treatment effects on gene expression and cell infiltration patterns in RA patients who did or did not respond to treatment.
From these and other results, the team came up with machine learning algorithms to predict rituximab or tocilizumab response or treatment refractory disease in individuals with RA based on the expression of a few dozen genes in synovial biopsy samples.
Past research suggests immune cell profiles and other features found in synovial biopsy samples from RA-affected joint tissue can help predict responses to targeted treatments, the authors explained. Even so, expression profiles are not typically used to guide RA treatment, despite the large proportion of patients with treatment-resistant or refractory forms of the disease.
"Approximately 40 percent of patients do not respond to individual biologic therapies and [5 to 20 percent] are refractory to all," they wrote, noting that "the exact mechanisms of response/nonresponse remain to be established."
To tease out some of those mechanisms and potential response markers, investigators performed semiquantitative immunohistochemistry profiling on pre-treatment biopsy samples from 28 individuals who went on to respond to rituximab treatment, and 54 RA patients who did not, as well as 37 tocilizumab responders and 42 tocilizumab non-responders.
Along with documented histological pathotypes for the cases, they did further histological profiling and RNA sequencing on pre- and post-treatment samples from dozens of rituximab- or tocilizumab-treated samples, as well as NanoString GeoMx digital spatial profiling on treatment responder, non-responder, and refractory samples.
With the help of analytical software known as glmmSeq, the researchers assessed gene expression patterns over time in the RA synovial biopsy samples, identifying more than 7,300 genes with elevated or reduced expression after treatment. Of those, just 345 genes were differentially expressed after treatment with both drugs.
While some synovial biopsy gene expression signatures coincided with response to both rituximab and tocilizumab, the team reported, other genes appeared to have altered activity in samples from responders to one drug or another. Together with histological features, for example, the gene expression data suggested that response to both targeted biologic drugs corresponded to humoral immune response shifts.
On the other hand, the researchers highlighted stromal- and fibroblast-related expression signatures shared by treatment-refractory RA patients who did not respond to either treatment, along with specific immune cell declines marking response to one drug or the other.
"The identification of genes and cell types associated with multidrug resistance could aid the development of new drugs for refractory patients in whom current medications targeting classical immune pathways are not effective," the authors concluded, explaining that biopsy-based analyses "could facilitate a patient-centered approach to the management of RA."