It is now well established that disease-modifying drugs (DMARDs), such as methotrexate (especially if used at higher weekly doses) or leflunomide, can significantly improve clinical and functional outcomes in RA and retard joint damage. This therapeutic success is limited because absence of active disease is rarely achieved. With the advent of biological agents, we have witnessed responsiveness in patients who were refractory to more traditional therapeutic measures. In addition, we learned that beyond increasing the proportion of responding patients it is possible to increase the magnitude of the response because many patients in clinical trials and clinical practice achieve a state of very low disease activity. Some even show no evidence of active disease.
More than a decade ago, response criteria were developed to allow for assessment of improvement of disease activity in the course of therapeutic interventions, particularly in clinical trials.1 These ACR response criteria call for a categorical minimal proportional improvement of disease activity from baseline rather than for the achievement of a categorical disease activity state. In fact, the discontinuous nature of the criteria, in conjunction with the variability of the baseline activity, does not allow for the determination of actual disease activity.
For clinical practice, however, criteria that allow evaluating actual disease activity may be more valuable for a variety of reasons. (See “Why We Need Disease Activity Criteria,” bottom right).
History of Assessment
The Disease Activity Score (DAS) and the 28-joint DAS (DAS28) were the first tools to provide a continuous disease activity measure for RA.7,8 The subsequently developed European League Against Rheumatism (EULAR) response criteria employed numerical changes of the DAS or DAS28 in combination with the achievement of a particular categorical disease activity state to determine a moderate or good response.9 This was an important step beyond assessment of mere disease activity improvement. Nevertheless, even patients with a good response are usually afflicted with residual disease activity, and even a response that leads to remission may only be called moderate if the numerical improvement criterion for a good response is not fulfilled.
With the use of biological therapies, the evaluation of response has reached a new dimension. Even patients who are non-responders by the traditional measures can have a virtual arrest of joint damage progression on a group level when treated with a TNF-inhibitor in combination with methotrexate, providing evidence for a dissociation of the inflammatory response with the destructive events by such therapy.10 Even with TNF-blocker combination therapy, radiographic changes progress more in high disease activity states than low disease activity states.11