Expert Perspectives on Advances in Precision Medicine in Treating Rheumatoid Arthritis - Episode 3

Tools for Improving Treatment Selection in RA

Published on: 

Vibeke Strand, MD, MACR, FACP, reviews the tools available to clinicians to improve treatment selection in RA, focusing on the molecular signature response classifier test.

Vibeke Strand, MD, MACR, FACP: I don’t think we have a lot of tools to tell us how to treat and what therapy. We know there’s the R4RA trial where synovial biopsies were performed, and RNA [ribonucleic acid] sequence-based identification showed that patients with low or absent B-cell lineage gene expression signature in their synovial tissue were significantly correlated with a better response with tocilizumab than rituximab. That makes a lot of sense, but we don’t have any more data from the R4RA for any other therapies. And there’s always the issue about synovial biopsy and obtaining those to get data. Unfortunately, synovial biopsies are something that we perform once. Most patients, once they’ve had one, aren’t willing to have a second one. So you aren’t able to follow those findings serially or to see what the effect is after 1 therapy. The only thing we have right now is the Scipher [Medicine] test, the MSRC [molecular signature response classifier]. It’s new and I don’t know that it’s widely used yet, but hopefully, it will be recognized as something to start with.

The MSRC is a molecular signature response classifier, and it is used for nonresponse to TNF inhibitors. In other words, who’s a patient who will not have a good result with a TNF inhibitor? It’s a tool that utilizes gene expression profiling, and that came from synovial biopsies. Again, it’s to predict the response of patients with RA [rheumatoid arthritis]. This therapy includes the gene expression data as well as microwave gene expression data, RNA sequencing, biologic network analyses, and machine learning. Essentially putting all this together in large patient cohorts has been able to identify a set of genes that are differentially expressed between those who respond to a TNF and the nonresponders, which are identified now by this test. It includes things like the genes, as I mentioned, also anti-CCP [cyclic citrullinated peptide] serologic status, age, sex, and BMI [body mass index]. This is made into a predictive algorithm. It’s numerically valued from 1 to 25. Those patients who have scores above 10.6 are considered to be nonresponders to TNF inhibitors. Overall, it’s a 23-feature molecular signature. In other words, it’s measuring 23 biomarkers that especially consider the pathways, both upstream and downstream, from what TNF does in rheumatoid arthritis.

Transcript edited for clarity