Blood Test Predicts TNF Inhibitor Response in RA Patients

Pam Harrison

November 16, 2014

A blood test that measures the ratio of interferon (IFN)-beta to IFN-alpha in patients with rheumatoid arthritis can help predict who will respond to tumor necrosis factor (TNF) inhibitors, according to a new study.

A higher IFN-beta/IFN-alpha ratio was strongly associated with a nonresponse to TNF inhibitors at 12 to 14 weeks in the study cohort.

"It depends on where you set the bar, but if you are looking for a really strong response to treatment, probably around half, or fewer, patients with rheumatoid arthritis achieve a 70% response to treatment," said Timothy Niewold, MD, from the Mayo Clinic in Rochester, Minnesota.

"A good percentage of patients don't respond to anti-TNF therapy even at a 20% response level," he told Medscape Medical News. "That was really the point of the study, because there really are people who do not respond significantly to these drugs, and we don't know who they are before we start prescribing."

The study results were presented at the American College of Rheumatology 2014 Annual Meeting in Boston.

Dr. Niewold and colleagues initially studied a test set of 32 patients with rheumatoid arthritis recruited from the Autoimmune Biomarkers Collaborative Network (ABCoN) registry.

All subjects had serum samples taken prior to treatment with a TNF inhibitor.

Before treatment, the investigators measured total serum type 1 IFN activity with a functional reporter cell assay, and compared IFN-alpha with IFN-beta activity.

"We found that the assay we've been using to measure interferon in lupus performs well in rheumatoid arthritis, and it's giving us a new window into circulating type 1 interferon in rheumatoid arthritis," Dr. Niewold explained.

The initial test set included only patients with a good response or no response to treatment, according to EULAR criteria.

An IFN-beta/IFN-alpha ratio above 1.3 in the serum prior to treatment with a TNF inhibitor was associated with a lack of response at 14 weeks (P = .009).

The investigators therefore chose 1.3 as the optimal cutoff point to test the usefulness of the interferon ratio as a predictor for nonresponse to TNF inhibitors.

For the validation set, the investigators recruited 80 additional patients with rheumatoid arthritis from the TETRAD registry. Serum samples taken before treatment with a TNF inhibitor were available for all patients.

In the validation set, patients with an interferon ratio above 1.3 were significantly more likely to have a nonresponse at 12 weeks (P = .0035), and no patient with a ratio of 1.3 or higher achieved a good response to treatment with a TNF inhibitor at 12 to 14 weeks.

The investigators performed a meta-analysis of the validation set, and found that the interferon ratio cutoff of 1.3 could be used to strongly discriminate between good responders and nonresponders at 12 to 14 weeks (odds ratio [OR], 18.0; P = 7.4 ×10–5).

And they found that the same cutoff could be used to discriminate between patients who did not respond to a TNF inhibitor at 12 to 14 weeks and patients whose response was moderate or good (OR, 4.34; P = .0014).

Our initial results are really promising

The mechanism of action of the TNF inhibitor — whether a monoclonal antibody or etanercept — did not influence this relation.

The assay used to measure the IFN-beta/IFN-alpha ratio is not yet ready for commercial use; more steps are needed before the test is widely available.

"But that's the ultimate goal," said Dr. Niewold. "We would still like to validate our findings in a larger set of patients before we can say for sure that the test is clinically applicable, although our initial results are really promising. The other issues are scale and the ability to perform the assay in an easier fashion."

This ratio might also be useful to predict response to TNF inhibitors in patients with other inflammatory conditions, such as Crohn's disease, he added.

A "Shotgun" Approach

The reality today is that rheumatologists do not have the equivalent of an antibiogram — which infectious disease physicians have — so that they can tell which antibiotic is going to work in a given infection, said Stephen Paget, MD, from Weill Medical College of Cornell University and the Hospital for Special Surgery in New York City.

"This means that we have to take a 'shotgun' approach and wait 2 to 3 months to see if these medications are going to work," he told Medscape Medical News. "And if they don't work, we put patients on another medicine."

If care could be individualized with specifically generated biomarkers that reflect the immunologic status of a patient before we prescribe a TNF inhibitor, "we could save a lot of money and give more focused care so that we don't waste the patient's time as their disease continues to progress," Dr. Paget explained.

"Of all the biomarker data that I've seen, I think this has tremendous importance with regard to who gets what medicine. As it is now, people will get two to three anti-TNFs before they are claimed to have failed TNF therapy, and that can take up to nine months," he added.

"If you could know upfront with a test like this that you could just as well avoid anti-TNF therapy in a given person, that would be massively important to the person, to the doctor, and to society as well," Dr Paget said.

Dr. Niewold reports financial relationships with Janssen Pharmaceutical Product LP, EMD Serono, and Biogen Idec.

American College of Rheumatology (ACR) 2014 Annual Meeting: Abstract 2927. To be presented November 18, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.