Q: How can autoantibodies and other biomarkers associated with RA development be used to intervene earlier in the disease process?

The prevention of preclinical RA is the holy grail of RA management, and, as we learn more, we are getting closer to achieving that goal. With respect to preclinical RA, however, there are currently 2 populations that should be considered separately: (1) those who do not meet RA criteria but have many of the symptoms and features of RA and are almost certain to meet the criteria in the near future, and (2) those who seem to be at high risk for developing RA but have few or even no joint symptoms and for whom, based on current technology, we are unable to predict whether they will, in fact, develop the disease. It is easier to address the former (ie, to identify and treat patients who have joint symptoms and synovitis but do not meet the clinical definition of RA). It makes intuitive sense that you could offer treatment (eg, methotrexate, abatacept) early and you would prevent some patients with preclinical RA from developing disease manifestations. In contrast, those with a high probability of developing RA based on family history and/or serology represent a greater challenge. If we could have more certainty about the inevitability of RA development, we could intervene earlier with greater confidence and prevent them from ever developing the disease.

Similar to cardiovascular risk factor reduction in heart disease, we are trying to advance the field of rheumatology so that we can identify and ameliorate RA risk factors and prevent the development of disease. Just like for the primary prevention of myocardial infarction or stroke, if we can identify the RA risk factors with sufficient precision to justify the risks of therapy, we can treat patients so that they never actually develop RA. RA is just as impactful as cardiovascular disease, but we have not been as successful as our cardiology colleagues in identifying those at highest risk, in part because, on average, RA is relatively uncommon (ie, approximately 1% of the general population). We also must frame the issue in ways that the patients understand. We have to be able to tell our patients in very clear terms about their risk of developing the disease and about what would likely happen to them if they get RA. They need to understand the implications of the disease. If we can develop a predicted risk score, it would allow us to give patients specific numbers and data that they can understand, and they may be more amenable to preventive treatments.

Being able to risk stratify patients to offer them the most appropriate, evidence-based therapy is key. For example, patients who have some joint symptoms but do not meet even the newer RA criteria (ie, the 2010 American College of Rheumatology/European League Against Rheumatism criteria) might have what we call undifferentiated arthritis. We are likely going to give these individuals methotrexate because they are likely going to develop RA. A trickier scenario would be patients who are anti-citrullinated protein antibody (ACPA) positive but have no joint symptoms at all. Up to a decade before people develop full-blown seropositive RA, they may have these autoantibodies (ie, ACPA and/or rheumatoid factor). In some of these patients, treatment might halt the development of the pathogenic processes triggering autoimmune and inflammatory disease, such that some may never develop RA. So, we need better risk stratification.

Other important questions are whether patients would find a decrease in the risk of developing RA to be an acceptable reason for starting preventive treatment and whether doctors would be willing to initiate these treatments. I agree that cardiovascular disease is ubiquitous and is familiar enough for people to understand the risks. But with other conditions that are less common and that most people are not as familiar with, such as RA, it may, frankly, be a tough sell, unless you have seen the disease before, perhaps in a family member. For example, even if you found a very strong and highly predictive marker (eg, a blood test) that could identify a subset of patients that had a 10- to 15-fold increased risk of developing RA, it still might not be that useful clinically. Since the prevalence of RA is, at most, 1%, a 15-fold increased risk would mean that you have a 15% chance of developing RA. Some patients will look at these data and realize that this also means that there is an 85% chance of not getting RA, and they will likely conclude that, because they feel fine, treatment is not warranted.    

There are several ongoing initiatives in the RA prevention space. These include the StopRA trial by Deane et al from the University of Colorado. The StopRA study is trying to identify patients who are at risk for RA and is looking at a variety of strategies for early disease prevention, including using risk factors such as family history and biomarkers such as ACPA and/or rheumatoid factor and following these patients over time. This trial is using hydroxychloroquine in a weight-based dosing regimen for 1 year. In addition, the PRAIRI study, conducted in the Netherlands, showed that a single infusion of 1000 mg of rituximab delayed the onset of disease compared with observation in high-risk, rheumatoid factor–positive, ACPA-positive patients with arthralgia. Finally, the ongoing APIPPRA study from the United Kingdom and the Netherlands is evaluating the efficacy of abatacept for preventing or delaying the onset of RA in very high–risk individuals.

We are also increasingly interested in RA risk modification—everything from smoking cessation to dental care, weight management, anti-inflammatory diets, and stress reduction. We want to offer something to high-risk patients who are ACPA positive and may have early signs of arthritis but do not meet the criteria for RA and are not able to enroll in a preventive clinical trial. Our first-line management strategy would likely be lifestyle modification. For those with a first-degree relative who has terrible arthritis and are willing to take medication, whether to prescribe pharmacologic treatment would be determined on a case-by-case basis until we get more data.