In the last 6 years, leflunomide, etanercept, infliximab, anakinra, and adalimumab have been approved by the Food and Drug Administration for the treatment of rheumatoid arthritis (RA). All of the tumor necrosis factor antagonists (etanercept, infliximab, adalimumab) have the ability to essentially halt radiographic evidence of progressive bony destruction (1–3). Responses according to the American College of Rheumatology (ACR) criteria (4) are uniformly substantial and clinically impressive when these agents are prescribed to patients who have achieved a less than total response to methotrexate. An alternative approach to adding a new disease-modifying antirheumatic drug (DMARD) is step-down treatment with prednisone, which may also have the ability to alter the course of radiographic destruction long term (5, 6).

The achievements associated with the introduction of these agents are meaningful in every sense. But can it now be said that the introduction of biologic agents has changed the outcome of the disease? How would this change be best measured and assessed? What level of evidence would be needed to convince even the most skeptical that these new treatments have resulted in a fundamentally different disease course?

The answers to these questions have very compelling practical implications. If it could actually be determined with certainty that the disease course of RA is altered with the use of these biologic agents, then it could also be posited that withholding the drugs that allow patients to achieve this improved outcome might be unethical. Are the therapeutic options that are presently available sufficiently powerful and safe to allow us to spread the word and officially announce that this near-miraculous state of affairs is already at hand? Is it indeed a professional, or even ethical lapse to not educate patients and medical colleagues if it can be established that major changes in disease outcomes have already occurred because of the use of these new agents?

Measurable changes in quality of life for patients may also gradually evolve in increments that are hardly apparent over a period of a year or less, but could become very obvious in retrospect over a period of several years or more. However, any interval of observation is somewhat arbitrary until sufficient time has passed to determine that negative outcomes, such as disability and disease-associated hospitalizations for joint arthroplasties, are being avoided when compared with historical norms. ACR response criteria or Disease Activity Score (DAS) responses by themselves may be inadequate to measure this outcome. As an analogy, consider an improvement in bone density measures without a reduction in fracture rates as inadequate to determine the actual utility of an agent used for the prevention of osteoporosis.

Conclusions about changes in the course of disease will also not necessarily apply to all patients. That is, it could be a dangerous oversimplification to state that all patients do well on new agents and that the course of disease is altered for all. In virtually all studies of the new DMARDs, ∼15–50% of patients are nonresponders according to the ACR 20% response criteria (2, 4, 7, 8). The challenge therefore is to spread the word, but make it defensible and evidence based. Also, optimal treatments will change, so views as to which drugs best improve the disease course must be flexible and evolve with the introduction of new agents.

One of the prime measures of whether disease course is altered with any intervention in patients with RA is avoidance of disability. Previously, long-term studies have demonstrated that a majority of patients with RA become disabled over periods as short as 6 years (9–11). Because none of the new agents have been widely used for periods that exceed that time frame, the determination of the rates of long-term disability compared with historical patterns will have to wait. There is also a potential problem with the way formal, long-term, open-label studies of drugs are performed because patients who have experienced a less-than-favorable response, or side effect, are no longer included in these types of cohorts.

Short-term investigations correct for this dropout phenomenon by using an intent-to-treat analysis, which corrects for early withdrawals for reasons that are not favorable to the intervention being studied by making certain that they are counted at the end of the study. However, real-world reporting systems may not make this kind of adjustment. Patients who change treatments are lost in the mix of other available treatments. Phase IV studies sponsored by the pharmaceutical industry are also unlikely to achieve the kind of precision needed to accurately assess disability as an outcome because patients may change treatments and be dropped from the study before they are disabled.

Therefore, to conclude that the fundamental course of RA has been altered, it should be demonstrated that both disability and orthopedic interventions can be avoided with a particular drug, or set of drugs and many of the present means of collecting data seem somewhat inadequate to that task. A database that collects long-term data from a wide range of patients on a variety of interventions would best serve the purpose of determining if disability can be avoided, while measuring traditional outcomes such as ACR response criteria and DAS scores. Quality of life should be measured along with frequency of drug side effects; development of comorbidities, including cardiovascular disease (12); survival; and overall patient and physician satisfaction. A few of these databases are in a maturing phase and have begun to yield useful data (13–15). Data collection systems that do not include physician-derived data have not been included in this discussion because of their inability to calculate ACR or DAS scores, which may be critical for the purpose of comparing and contrasting different methods used to determine if the course of disease is altered. Data from a single academic practice indicate that typical joint counts are dramatically reduced from traditional norms (16); these observations need to be confirmed in larger, more diverse populations of patients from multiple sites.

Table 1 summarizes the difficulties inherent in establishing with certainty that the course of RA has been changed, along with suggested approaches to the problems. Fortunately, many of the impediments to establishing this conclusion will be removed as large prospective databases collect sufficiently detailed information to address all of these questions. However, if decreased mean joint counts, significantly improved ACR responses and DAS scores, and favorable radiographic outcomes are relevant, then it appears that there is at least a signal that a true change for the better in the course of RA is indeed becoming a reality. The extent to which theses changes in disease can be documented in large real-world populations will depend upon the successful adoption and dissemination of large electronic databases that are presently beginning to be used.

Table 1. Difficulties and solutions with establishing the occurrence of a fundamental change in the course of rheumatoid arthritis (RA)
 Insufficient duration of disability data associated with new medications.
 Insufficient comparator data with historical norms on quality of life data such as The Health Assessment Questionnaire.
 Insufficient data on frequency of comorbidities such as cardiovascular disease and premature death from current interventions in comparison with historical data, where the same problem of insufficient data exists.
 Insufficient data on the frequency of orthopedic interventions with presently used interventions when compared with historical data.
 Insufficient data on the cost:benefit ratio to patients and society of present interventions over sufficient periods of times to reach defensible conclusions.
Possible solutions
 Long-term, physician-derived databases gathering data on relevant outcomes to address the questions concerning disability, death, cardiovascular disease, hospitalizations for orthopedic procedures, joint counts, physician assessment of disease activity, physician ascription of causality of toxicities and comorbidities with interventions, radiographic outcomes, and bone density scores. All of these factors should be corrected in analyses for disease duration, disease severity as judged by both physician and patient, age, family history, smoking history, prednisone use, and intraarticular corticosteroid use.
 Utilization of electronic databases within the rheumatology community to document and disseminate outcomes.
 Clearly defined benefits to physicians to adopt electronic databases and share data.


  1. Top of page
  • 1
    Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH, Keystone EC, et al. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis [published erratum appears in N Engl J Med 2001;344:240 and N Engl J Med 2001;344:76]. N Engl J Med 2000; 343: 158693.
  • 2
    Lipsky PE, van der Heijde DM, St. Clair EW, Furst DE, Breedveld FC, Kalden JR, et al, and the Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. Infliximab and methotrexate in the treatment of rheumatoid arthritis. N Eng J Med 2000; 343: 1594602.
  • 3
    Keystone EC, Kavanaugh AF, Sharp JT, Tannenbaum H, Hua Y, Teoh LS, et al. Radiographic, clinical, and functional outcomes of treatment with adalimumab (a human anti–tumor necrosis factor monoclonal antibody) in patients with active rheumatoid arthritis receiving concomitant methotrexate therapy: a randomized, placebo-controlled, 52-week trial. Arthritis Rheum 2004; 50; 140011.
  • 4
    Felson DT, Anderson JJ, Boers M, Bombardier C, Furst D, Goldsmith C, et al. American College of Rheumatology preliminary definition of improvement in rheumatoid arthritis. Arthritis Rheum 1995; 38: 72735.
  • 5
    Boers M, Verhoeven AC, Markusse HM, van de Laar MA, Westhovens R, van Denderen JC, et al. Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with sulphasalazine alone in early rheumatoid arthritis [published erratum appears in Lancet 1998;351:220]. Lancet 1997; 350: 30918.
  • 6
    Landewe RB, Boers M, Verhoeven AC, Westhovens R, van de Laar MA, Markusse HM, et al. COBRA combination therapy in patients with early rheumatoid arthritis: long-term structural benefits of a brief intervention. Arthritis Rheum 2002; 46: 34756.
  • 7
    Weinblatt ME, Kremer JM, Bankhurst AD, Bulpitt KJ, Fleischmann RM, Fox RI, et al. A trial of etanercept, a recombinant tumor necrosis factor receptor:Fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. N Engl J Med 1999; 340: 2539.
  • 8
    Kremer JM, Genovese MC, Cannon GW, Caldwell JR, Cush JJ, Furst DE, et al. Concomitant leflunomide therapy in patients with active rheumatoid arthritis despite stable doses of methotrexate: a randomized, double-blind, placebo-controlled trial. Annals Int Med 2002; 137: 72633.
  • 9
    Barrett EM, Scott DG, Wiles NJ, Symmons DP. The impact of rheumatoid arthritis on employment status in the early years of disease: a UK community-based study. Rheumatology (Oxford) 2000; 39: 14039.
  • 10
    Sokka T, Kautiainen H, Mottonen T, Hannonen P. Work disability in rheumatoid arthritis 10 years after the diagnosis. J Rheumatol 1999; 26: 16815.
  • 11
    Wolfe F, Hawley DJ. The longterm outcomes of rheumatoid arthritis: work disability: a prospective 18 year study of 823 patients. J Rheumatol 1998; 25: 210817.
  • 12
    Del Rincon I, Williams K, Stern MP, Freeman GL, Escalante A. High incidence of cardiovascular events in a rheumatoid arthritis cohort not explained by traditional cardiac risk factors. Arthritis Rheum 2001; 44: 273745.
  • 13
    Van Vollenhoven RF, Klareskog L. Clinical responses to tumor necrosis factor α antagonists do not show a bimodal distribution: data from the Stockholm tumor necrosis factor α followup registry. Arthritis Rheum 2003; 48: 150003.
  • 14
    Eyre S, Barton A, Shephard N, Hinks A, Brintnell W, MacKay K, et al. Investigation of susceptibility loci identified in the UK rheumatoid arthritis whole-genome scan in further series of 217 UK affected sibling pairs. Arthritis Rheum 2004; 50: 72935.
  • 15
    Kremer JM, Lavelle L, Hinkle K, Reed G. Clinical factors associated with infection, and opportunistic infection, in patients with rheumatoid arthritis (RA) from the CORRONA database [abstract]. Arthritis Rheum 2004; 50 Suppl 9: S688.
  • 16
    Sokka T, Pincus T. Eligibility of patients in routine care for major clinical trials of anti–tumor necrosis factor α agents in rheumatoid arthritis. Arthritis Rheum 2003; 48; 3138.