Q: How do you treat patients with RA who have persistent moderate disease activity despite receiving biologics/targeted immunomodulators?

With the advent of treat-to-target approaches, there has been much emphasis placed on getting patients to a state of low disease activity (LDA) or remission. We have many well-known metrics that we use to evaluate disease activity and reach targets, such as the Clinical Disease Activity Index, the Disease Activity Score 28 with C-reactive protein (CRP), and the Routine Assessment of Patient Index Data 3. However, approximately one-third of patients do not achieve what the doctor or the patients want in terms of their treatment goal(s), despite having an extraordinarily rich pipeline of drugs available. One of the leading reasons for failing to achieve LDA or remission is the presence of elevated disease activity based on the patient’s global response. The tender/swollen joints and laboratory values may be in a good range, but factors such as pain and fatigue are causing patients to fail to achieve LDA or remission, and, so, patients may not be feeling as well as they would like. In these cases, we need to drill down to better understand these patients and what the deficits are. For those who are not at target, we must attempt to determine whether they have persistent molecular disease activity. Alternatively, the data may suggest that the patient is suffering from more subjective—but real and meaningful—deficits of symptom control. Swollen joints are more important than tender joints in this setting because those who have central pain amplification tend to have more tenderness. Additionally, we rely on objective findings such as the erythrocyte sedimentation rate (ESR) and CRP. We also have imaging modalities (eg, magnetic resonance imaging, ultrasound) and molecular markers for disease activity (eg, the multi-biomarker disease activity [MBDA] test). Thus, there are many tools available to measure the inflammatory molecular response to therapy; it is the other part that is more challenging.

Despite the current availability of many effective therapeutic agents, there is a subset of patients who do not respond at all or who do not respond adequately to any given therapy. There are many components to that. It is clear that people respond best to the first thing that you give them, and they respond less and less as you move down the treatment algorithm. We do not know whether this identifies a population that is difficult to treat or whether the disease changes over time in response to therapy, but it increases the importance of selecting the right therapy up front. However, it is a struggle to identify the biologic that is the most likely to work. If the first biologic fails (typically an anti–tumor necrosis factor [anti-TNF] agent), we do not know whether we should switch to another anti-TNF agent or to a different class of drug. Another issue is the definition of a nonresponder. We have always presumed that it is the inflammatory arthritis that is not responding. But we are increasingly recognizing that some of the nonresponse is a different animal. This is the issue of central noninflammatory pain. If you have a patient who is not responding to a biologic and they have pain but little inflammation, that is a different kind of nonresponse. Switching to another biologic in these individuals may not be the answer because it is the noninflammatory component that needs to be controlled. For these patients, we use other approaches, including physical therapy, exercise, good sleep hygiene, and medications such as duloxetine and pregabalin. Another challenging subset of patients are those whose inflammation is not controlled with the first biologic. For primary failures to an initial anti-TNF agent, switching to another class of drug may be better. For failures that occur secondarily (after an initial response), a second agent in the same class may be appropriate. But we need tools such as biomarkers to guide treatment selection.

It is important to make sure that you are treating the right thing. For patients who have established RA, there is the potential that they will conflate other symptoms (eg, joint pain from osteoarthritis, mechanical lower back pain) with active RA. I agree that traditional biomarkers of inflammation (eg, CRP, ESR) and/or the MBDA as well as imaging can be used to ensure that the reports of persistent pain are due to active synovitis and active RA and not to chronic pain from another cause. Patient-reported outcomes are also informative in this regard in that symptoms of depression, anxiety, sleep problems, and fatigue may be clues when trying to tease apart what is active RA and what is not. It is also helpful to set expectations up front so that patients understand that switching biologic or targeted therapies is part of the treatment course. We tell them that we will be changing medications if they do not achieve at least LDA, and that helps to mold their expectations. It gives patients greater confidence in your knowledge and abilities and helps them to understand that switching treatments is not a consequence of vacillating or a lack of certainty. Regarding those who fail biologic therapy, there are not many who are true primary failures. For patients who do not feel better on an anti-TNF agent, I worry whether I am treating the right thing. If you fail the first biologic because of a safety/tolerability issue, you are more likely to respond to a second agent than if you had failed the first agent because of lack of efficacy. Although switching to an agent with a different mechanism of action has been shown to increase efficacy, the absolute difference in the magnitude of the benefit is small. Thus, the preponderance of evidence likely gives a small tip of the hat to changing mechanism of action rather than to anti-TNF cycling.