22 April 2021

Study highlights challenges for baseline prediction of TNFi response

Rheumatoid Arthritis TNF

A small study of RA patients taking adalimumab has identified gene expression pathways that change in response to positive treatment.

However, the study did not demonstrate any clear difference at baseline between patients who would respond to TNF-inhibitor therapy and those who failed treatment.

“There are some changes [to gene expression] there, but they don’t seem to be obvious at baseline,” said Dr Simon Chatfield, a rheumatologist at the Royal Melbourne Hospital who completed his PhD on inflammatory markers in arthritis at the Walter and Eliza Hall Institute of Medical Research.

The findings therefore won’t be making any strong impact on clinical choices for the time being. Rather, Dr Chatfield told Rheumatology Republic, the study highlights an area of understudied research and generates new hypotheses to investigate further.

The UK study selected seventy biologic-naïve patients, previously treated with DMARDs, from the Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate, who could be categorised as ‘good’ or non-responders after 3 months on adalimumab.

Blood samples were collected pre- and post-treatment, with gene expression levels measured using a transcriptome-wide panel to profile early responders to adalimumab and those who didn’t respond after 3 months.

Though the study aimed to discover and validate biomarkers of treatment response, the results at most have provided an insight into the genetic mechanisms underpinning response to adalimumab.

Gene expression levels differed between baseline and 3-months follow-up in good responders only. These changes, identified in over 800 transcripts from gene pathways linked to TNF, were subsequently validated in an independent cohort of another 22 RA patients using a secondary technique that directly measures RNA transcript levels.

“This study identified many immune-related genes with increased expression in the whole blood of good-responders,” the University of Manchester team wrote in their paper.

“Whilst seemingly paradoxical, this may represent migration of RA-associated inflammatory factors out of the affected joint and into the peripheral blood in a positive response to TNFi therapy,” said the study authors.

However, said Dr Chatfield, those changes are only really apparent comparing patients who did respond to treatment, from baseline to three months.

“The main question is can we tell at baseline who is going to respond or not, and in this study, they couldn’t find a difference,” he said.

No baseline biomarkers predictive of treatment response were identified, and the study failed to observe any significant difference between adalimumab responders and non-responders at 3 months.

The study authors reasoned that this was due to their small discovery cohort which contained more responders than non-responders, 50 and 20 patients respectively.

Dr Chatfield also noted that the study’s patient selection, whilst clear, makes interpreting the findings in a clinical setting difficult. “They’ve taken patients at the extremes [of treatment response] rather than splitting them down the middle,” he said.

Patients were classed as good responders if their swollen joint count was less than 2.6 after 3 months on treatment; non-responders were those who had minor improvements on their DAS28 score with an endpoint over 5.1.

Altogether, Dr Chatfield said the study findings reflect that “our ability to predict responders at baseline, by any means – clinical, laboratory or radiologically – is still pretty limited.”

“It’s incredibly hard to identify biomarkers of any description that are really robust in predicting something that should be biologically distinct,” he said.

Arthritis Research & Therapy 2021, 10 March