LAS VEGAS — A new diagnostic algorithm has been developed to simplify strategies to manage difficult-to-treat rheumatoid arthritis (D2T RA), an umbrella term for persistent symptoms after multiple trials of standard therapies. 

The key point is to “think about phenotype before jumping into management,” according to its author, Bryant R. England, MD, PhD, chief of the Division of Rheumatology and Immunology at the University of Nebraska Medical Center in Omaha.

The algorithm is simple because there are only three decision points. Regardless of the many terms used to describe D2T RA, such as refractory RA, treatment-resistant RA, and poly-refractory RA, the main role of the decision tree is to separate symptoms likely to be driven by inflammation from those that are not. 

“Careful phenotyping will tell you the management pathway. It is the most important thing you can do,” said England, who outlined his algorithm at Perspectives in Rheumatic Diseases (PRD) 2026. 

Patients Confound Symptom Sources

Patients who present with a sizeable list of nonspecific complaints, such as functional decline, chronic pain, and fatigue, often do not try or even think to distinguish which of them stems from their arthritis. England said such patients sum up their health status as “It’s bad, doc” with the explicit message that they are depending on the physician to figure out why.

But comorbidities can contribute to traditional symptoms of RA, such as impaired mobility, while RA can contribute to symptoms not necessarily directly related to joint disease, such as fatigue. Phenotyping patients with a profile of D2T RA allows the clinician to more clearly consider the sources of symptoms, some of which have little or nothing to do with joint disease, England said.

“Not all D2T is refractory inflammatory RA,” said England, emphasizing the main point of his remarks.

The algorithm that England proposed builds on EULAR D2T RA recommendations published in 2022. The EULAR definition of D2T RA is fulfilled by three criteria: failure on at least two biologic or targeted disease-modifying antirheumatic drugs (DMARDs); signs of continued active and progressive RA; and persistent symptoms considered problematic by the clinician or the patient.

The EULAR recommendations are comprehensive, covering a broad array of factors that might contribute to an inadequate therapeutic response, including a misdiagnosis and poor adherence. In England’s more-simple algorithm, he is far more focused. 

Using cases to illustrate D2T RA phenotypes, England provided three examples. The first was the “chronic pain or pain sensitization” phenotype, a presentation that typically involves pain out of proportion with inflammatory features.

Multimorbidity Phenotype Symptoms Are Mixed

The D2T RA “multimorbidity” phenotype is characterized by concomitant pathologies, which typically contribute to or cause many of the nonspecific features, such as fatigue and impaired mobility.

The most common inflammatory phenotype, “persistent inflammatory refractory RA,” which England identified by the acronym PIRRA, captures patients who are simply inadequately responsive to standard therapies. While this is the phenotype most closely associated with D2T RA, England believes inflammation participates in only about half of D2T RA cases, meaning noninflammatory causes deserve equal attention.

This might explain why estimates of the prevalence of D2T RA “are all over the map,” he said. While England reported the accepted figure of D2T RA is about 12% based on systematic reviews of the literature, individual studies have come up with figures as low as 5% and as high as 25%. There are many potential explanations, but differences in how noninflammatory causes are counted might be among them.

The first step for a patient complaining of uncontrolled symptoms on a follow-up visit is to consider the likelihood of an RA-related cause, according to England’s algorithm. Ultimately, this can require evaluating disease activity by standard measures, such as swollen joint count and systemic inflammatory biomarkers, but clinicians should be alert to phenotypes characterized by excess pain or comorbidities even during the course of this workup.

Key to this workup is determining whether returning patients whose symptoms were once controlled are still on target, according to England.

This step can rule D2T RA in or out. Patients considered to have D2T RA can be moved further down the algorithm to look for causes. Those who are not are sent to a workup of noninflammatory reasons for symptoms. 

An illustration of how this works was provided by a theoretical case in which characteristics and symptoms were tweaked to describe three D2T RA phenotypes. The case was a 63-year-old woman who returned to the office with complaints of pain in specific joints and expressed concern her therapy is no longer working. 

In the first case, the patient did not have systemic inflammation on the basis of laboratory tests despite some tender joints. In the absence of evidence of disease activity, the likely explanation for her stiffness was concomitant spinal disease and other comorbidities that kept her largely at home. In other words, the etiology was not inflammatory and would not be expected to improve with adjustments in her anti-inflammatory therapies.

Joint Complaints Are Not Always Due to RA

The second case was similar, but she was even more compromised by a long list of comorbidities that included coronary artery disease, congestive heart failure with a low ventricular ejection fraction, chronic obstructive pulmonary disease, and osteoarthritis. Again, although the patient did have some tender joints, no strong evidence of active inflammation was found in the evaluation. There was no indication in this patient, like the first, for adjusting the RA therapy.

The third case presented with obesity but had far more extensive joint involvement than the others. Both the tender joint count and the swollen joint count was 14. The hands of the patient were so swollen that she could not make a fist. Systemic inflammation was confirmed on laboratory testing. This patient, unlike the others, would be moved to the next decision point.

This decision point is now limited to patients with established D2T RA. The next step involves determining whether the RA is refractory to treatment or is being complicated by other factors, such as poor adherence or pain without inflammation. Those with persistent RA-related manifestations despite optimized therapy can be moved down the algorithm to the last of the three decision points. 

But this decision point, like the other two, still involves separating inflammatory from noninflammatory causes. 

At this step, even if patients have active inflammation and progressive disease on optimized DMARDs, noninflammatory factors still have to be considered. D2T RA can be complicated in patients with multiple comorbidities, and all contributing factors should be considered. 

These concomitant clinical issues are broad, but England recommended a shorthand for quickly organizing the possibilities. His approach involves assigning each finger to a group of disorders, some of which can be treated nonpharmacologically.

The thumb was assigned to cardiovascular health, often improved with smoking cessation, increasing physical activity, and attaining a healthy weight. The remaining four fingers starting from the index finger were assigned to cancer, respiratory disease, infection, and osteoporosis, respectively. 

He added his palm as a reminder to consider mental health. In this case, depression and inadequate sleep, like health risks in the other categories, can diminish a health status that is relevant to well-being and can be contributing factors in D2T RA.

D2T RA Is Only Sometimes Treatment Refractory 

Ultimately, treatment-refractory RA at the biologic level is not the first nor the only cause to consider in D2T RA, according to England. With phenotyping in the context of his algorithm, he believes that a large proportion of D2T RA cases can be addressed.

Jacob “Jaap” M. van Laar, MD, PhD, a professor of rheumatology and clinical immunology at the University Medical Center in Utrecht, Netherlands, has written extensively on D2T RA. The senior author of the 2022 EULAR D2T RA recommendations as well as a more recent 2025 review article, he said that the algorithm proposed by England is consistent with his own approach.

“Management of RA, and D2T RA patients in particular, requires a holistic approach, always,” he said. 

This concept of attempting to address D2T RA by inflammatory and noninflammatory causes might seem obvious, but van Laar thinks the message has not been adequately circulated.

“I have seen many D2T RA patients suffer due to inappropriate focus on intensification of antirheumatic medication rather than common comorbidities such as obesity and fibromyalgia,” he said. Speaking of the concept outlined by England, he added, “I am very pleased to see this notion being picked up.”

England reported financial relationships with Artiva Biotherapeutics and Boehringer Ingelheim. Van Laar reported financial relationships with AbbVie, Alfasigma, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly, Galapagos, GlaxoSmithKline, and Novartis.

Ted Bosworth, a career medical writer based in New York City, has been covering advances in clinical medicine, including rheumatology, for several decades.