Diagnosis of Early Rheumatoid Arthritis

Introduction

The earliest possible diagnosis and treatment of nascent rheumatoid arthritis (RA) is the goal for all clinicians who care for patients with musculoskeletal complaints. A growing body of evidence has underscored the consistent clinical and radiographic benefits of early aggressive treatment and the unfortunate consequences of either delayed or ineffective treatment. Early diagnosis is a challenge to physicians and all health care systems. This tenet is essential to RA management and attested to by other disciplines wherein the earliest identification of a pathologic process has been fundamental to the management of neoplastic, infectious, neurologic and developmental disorders. In all of these, early recognition increases the odds of optimal outcomes, largely by altering the initial trajectory of disease.

Population-based studies of disease-modifying antirheumatic drug (DMARD)–treated patients have consistently shown that RA patients are at substantial risk for progressive joint damage, disability and increased morbidity and mortality. Such adverse sequelae are likely if potent disease-modifying therapies are avoided or delayed. Numerous studies have demonstrated the importance of the earliest possible referral, diagnosis and initiation of appropriate DMARD therapy in patients with RA. Importantly, early diagnosis and reduction in the lag time to DMARD initiation has been shown to yield significant future clinical, functional and radiographic benefits. Studies have attested to the concept that there is a window of opportunity when treating RA patients.

A Window of Opportunity

The focus on an earlier and potentially modifiable disease state questions whether a therapeutic window of opportunity exists. This chronologic construct identifies a time frame wherein intervention yields the greatest impact on disease progression. The goal of aggressive time-targeted intervention then would either be true remission, therapeutic remission, or at least a halting of disease progression as measured by functional and radiographic change. The ominous nature of this window is supported by magnetic resonance imaging (MRI) and radiographic evidence of erosive damage in the first 4 months of early RA. Moreover, an analysis of data from the French ESPOIR and the Dutch Leiden Early Arthritis Clinic cohorts found a non-linear relationship between symptom duration and remission hazard ratios, consistent with the existence of an early therapeutic window.

The definition of early RA has varied over the last decade but clearly the window of opportunity has narrowed with time. In the pivotal anti-TNF early RA trials (Early Rheumatoid Arthritis [ERA], Active-Controlled Study of Patients Receiving Infliximab for Treatment of Rheumatoid Arthritis of Early Onset [ASPIRE], Prospective Multi-Centre Randomized, Double-blind, Active Comparator-Controlled, Parallel-Groups Study Comparing the Fully Human Monoclonal Anti-TNFα Antibody D2E7 Given Every Second Week With Methotrexate Given Weekly and the Combination of D2E7 and Methotrexate Administered Over 2 Years in Patients With Early Rheumatoid Arthritis [PREMIER] trials), the disease duration (from diagnosis to enrollment) was limited to 3 years and the observed mean disease duration was ≤12 months in these studies. While many would have once defined early RA as having a disease duration of <3 years, Aletaha and colleagues’ survey of rheumatologists showed that over time, this perceived window of opportunity has narrowed.

Most rheumatologists now define “early RA” as <3 months of symptoms and disease activity. This time frame is supported by several studies showing that the earlier the intervention, the better the outcome. Nell and associates compared the outcome of early RA patients (median disease duration, 12 months) with those with very early RA (VERA) (median duration, 3 months). These patients were treated with conventional DMARDs (MTX, sulfasalazine [SSZ], antimalarials, leflunomide, cyclosporine, etc) and, after 36 months, superior responses were seen in the VERA patients. VERA patients exhibited greater ACR20 (70% vs 40%), ACR70 (55% vs 20%) and disease activity score (DAS) improvements (2.8 vs 1.7). Moreover, the VERA patients had significantly less radiographic progression as measured by Larsen scores (Figure 6-1). Thus, treatment of early RA may best be defined as “the earlier the better.” While American College of Rheumatology (ACR) criteria for RA require 6 weeks of symptoms and studies show that 12 weeks of symptoms best define persistence or chronicity, inclusion in early RA studies will require at least 6 to 12 weeks of symptoms. However, clinicians should not be bound by such time constraints since research has shown that treatment should be initiated as soon as symptoms are documented. Moreover, therapy should complement the severity and persistence of the synovitis rather than waiting to reach a time-constrained diagnosis.

Enlarge 

Preclinical RA

Many patients will present with arthralgias or insufficient number of swollen joints, such that they can not be diagnosed with RA or fulfill the ACR/EULAR criteria. Such patients may be labeled as undifferentiated polyarthritis, pre-clinical RA, or clinically suspect arthralgia; the distinction between these subsets is not entirely clear. Patients with early undifferentiated arthritis are common. Observational studies have shown that 40% to 50% of undifferentiated polyarthritis patients will go into remission within 2 years and that roughly one third may evolve into RA. Many of these patients respond well to a course of glucocorticoid therapy. Not knowing whether the undifferentiated arthritis patient with inflammatory synovitis will progress or remit poses a management dilemma—should DMARD treatment be started or delayed until a definitive diagnosis can be made? The Probable Rheumatoid Arthritis Versus Placebo Therapy (PROMPT) trial has shown us that aggressive use of MTX in undifferentiated arthritis patients provides a short-term delay in progression to RA. The PROMPT trial strongly supports the concept of a window of opportunity in RA.

Researchers from the Leiden Early Arthritis Clinic followed 570 early undifferentiated arthritis patients and developed a decision tool that predicts who will develop RA. From this cohort, 177 went on to develop RA and 150 went on to a drug-free remission. Nine common clinical variables from the patient’s first visit (Figure 6-2) were shown to predict those who would develop RA. Predictive variables (with possible scores of 0-14) include age, sex, joint distribution, morning stiffness, the number of tender and swollen joints, C-reactive protein

(CRP), rheumatoid factor (RF) and cyclic citrullinated peptide (CCP) antibodies. Patients with a score <3 never developed RA and those with a score of ≥11 always developed RA. As the prediction score rose from 4 to 10, so did the odds of developing RA. For example, 91% of patients with undifferentiated arthritis who had a score ≤6.0 did not develop RA (negative predictive value 91%) and for those with a score ≥8.0, 84% progressed to RA (positive predictive value 84%). While this tool overstates the obvious (the more RA features, the more likely RA develops), it may be useful in quantifying who should not receive aggressive DMARD therapies (e.g., those with scores of <6). Unfortunately, the calculation of a prediction score is not clinic-friendly, thus limiting the utility of this tool to clinical trials or computer-based algorithms.

Pre-clinical RA (see Preclinical Rheumatoid Arthritis) has been defined as patients with chronic polyarthralgia, who are often seropositive for anticitrullinated protein antibody (ACPA) or RF antibodies, but who do not exhibit palpable synovitis. In some studies, these patients are said to have “clinically-suspect arthralgia” (CSA). Seropositive pre-clinical RA patients have a roughly 30% odds of progressing to chronic inflammatory polyarthritis (RA). Studies have shown the odds of progression to RA are increased in first-degree relatives of RA probands, those with very high titers of RF or ACPA, those who are double-positive for RF and ACPA, smokers and those exposed to environmental pollutants, people with obesity, and those with imaging evidence of synovitis or tenosynovitis. The pathogenesis of pre-clinical RA suggests that a genetically susceptible host (with the “shared epitope” or PTPN22) becomes environmentally challenged (smoking, pollution, periodontal disease, obesity), leading to an autoimmune state (RF or ACPA positivity) that with time may immunologically evolve into clinically manifest rheumatoid inflammatory polyarthritis.

Many clinical trials have tested the efficacy of specific therapeutic interventions in patients with pre-clinical RA or CSA (see Preclinical Rheumatoid Arthritis for more detailed discussion). There are numerous well-done randomized trials have failed to show that treatment of preclinical RA with either rituximab (PRAIRI study), methotrexate (PROMPT and TREAT-EARLIER trials), hydroxychloroquine (STOP-RA trial), or atorvastatin (STAPRA),could significantly delay or prevent the onset of RA. By contrast, several trials of abatacept (ADJUST, ARIAA, APIPPRA) have shown that abatacept can either delay or prevent RA onset in the short-term (12-24 month).

Nonetheless, most patients with undifferentiated polyarthritis, CSA, or pre-clinical arthralgias are treated symptomatically, without DMARDs or biologics as there is no clear consensus as to which patients will benefit from more aggressive management.

Enlarge 

Access to Care

Whereas early diagnosis of RA may be fraught with obstacles, patient access to rheumatologic consultation and care is limited by several factors. The Centers for Disease Control (CDC) has estimated that >10 million Americans have chronic joint symptoms but have never seen a health care provider. While many of these will have degenerative arthritis, inflammatory etiologies are likely to be included as well. In the United States, there is a large RA population (>1.3 million), many of whom have not been diagnosed or treated.

Furthermore, many of patients who have been diagnosed are cared for by primary care physicians (PCPs) who are faced with the challenge of diagnosing and managing this complex disorder. PCPs are more likely to see patients in the first few weeks of their illness, when few joints are affected, laboratory abnormalities may be absent and radiographs are normal. This crucial stage of disease is also when remission and optimal clinical responses are achievable, provided appropriate and aggressive therapy is initiated.

Access to rheumatologic consultation is problematic for many PCPs and patients. With <5,000 practicing rheumatologists in the United States, there is a bottleneck for rheumatology referrals for an estimated 130,000 new cases of RA that develop each year. By 2030, the demand for rheumatologists is estimated to exceed supply by 102%. This unmet need will require better education of PCPs and dedicated efforts by rheumatologists to facilitate the referral of patients with new-onset disease (see Preclinical Rheumatoid Arthritis).

Impact of Intervention

The benefits of early intervention have been well documented. A meta-analysis of >1400 patients in 14 randomized controlled trials of conventional DMARDs underscored the importance of early DMARD use (Table 6-1). Regardless of the DMARD used, clinical responses were consistently best when it was used earlier in the disease. Several studies have also shown the lag time from symptom onset to DMARD use to be critically important. Lard and colleagues reported distinctly different outcomes for two cohorts with early disease (mean disease duration, 4 to 5 months). Whereas patients given early (median, 15 days) DMARD therapy (hydroxychloroquine (HCQ) or sulfasalazine (SSZ)) had negligible radiographic change after 2 years, those with a delay (median, 123 days) in DMARD initiation demonstrated significant radiographic progression over 2 years (Figure 6-3). A 4-year follow-up study of these same patients showed that between the second and fourth years, both groups progressed equally; however, those with no DMARD delay still fared better over time.

Several other studies have also documented the long-term, downstream benefits of earlier intervention. In the Combinatietherapie Bij Reumatoide Artritis (COBRA) trial (see Conventional Synthetic Disease-Modifying Antirheumatic Drugs), 155 DMARD- and prednisone-naïve patients with RA of <2 years’ duration (mean, 4 months) were randomized to receive SSZ monotherapy or combination therapy with SSZ, MTX and high-dose prednisolone. After 6 months of therapy, the combination group had significantly higher ACR20 (72% vs 49%), ACR50 (49% vs 21%) and more remissions (28% vs 16%) than the SSZ-alone group. When the corticosteroids and MTX were withdrawn, the clinical benefits were nullified. Nonetheless, long-term follow-up studies have demonstrated that those initially treated with aggressive combination therapy were more likely to have less radiographic damage and less disability over 5 years (Figure 6-4).

The Finnish Rheumatoid Arthritis Combination Therapy (FinRA-Co) trial was a 2-year, open-label study in 195 patients with early RA (<2 years). Patients were treated with either sulfasalazine monotherapy or triple-DMARD therapy (MTX, SSZ, HCQ) and all were allowed prednisone 5-10 mg per day. After 2 years, better results, including greater remission rates (38% vs 18%), were seen in those treated with early aggressive combination DMARDs compared with monotherapy. Radiographic scores worsened from 2 to 12 Larsen Units in the monotherapy group but only from 2 to 4 Larsen Units in the triple-DMARD group. Interestingly, a subanalysis of the SSZ monotherapy group showed that remission rates were significantly lower for those experiencing a DMARD treatment delay of >4 months (11% vs 35% in those with no DMARD therapy).

While the COBRA study showed a cost-efficacy trend favoring early combination DMARD therapy, the FinRA-Co Study was able to show the long-term benefits of early, aggressive combination therapy with significantly less work disability, premature retirement and sick leave for those initially treated with triple DMARD therapy. Results from 661 patients in the French prospective observational ESPOIR cohort demonstrated a significantly lower mean 12-month radiographic progression score in patients who started DMARD therapy within 3 months of diagnosis compared to those who started later (P = 0.033). These are but a few example studies that show the long-term symptomatic, functional and radiographic benefits of early DMARD or combination DMARD therapy in RA patients. TNF-Blocking Therapies reviews in detail the outcomes of early RA patients treated with MTX combined with etanercept, infliximab, or adalimumab. These studies demonstrate how early, aggressive anti–TNF therapy can also lead to superior clinical and radiographic outcomes.

All told, these data suggest a recipe for success in RA therapy:

• Early diagnosis leads to early therapy
• The earlier the DMARD initiation, the better
• Delays in DMARD initiation can have disastrous downstream effects on clinical, radiographic, disability, and cost outcomes
• The choice of the first DMARD (single or combination therapy) may be the most important treatment decision
• Combination DMARDs (especially MTX plus TNF inhibitors) offer the best chance for remission and a halting of radiographic progression in this progressive disease.

Enlarge 

Enlarge 

Enlarge 

Who Should Be Referred?

Ideally, all patients with recent-onset RA or inflammatory arthritis should be evaluated by a rheumatologist. Diagnostic certainty may be fleeting in early disease since the symptoms and outcomes of new-onset arthritis vary. Moreover, many EACs have shown that a high percentage of new-onset synovitis will either remit or evolve into other acute or chronic inflammatory arthritides. Yet, others who are seropositive are at a significant risk to evolve into RA. Thus the challenge is to identify those clinical clues that can either enhance diagnostic certainty or be used as rules for referral and consultation with a musculoskeletal specialist.

Green and colleagues examined a cohort of 63 patients with early RA and found that persistent symmetric disease for ≥12 weeks was the best predictor of persistent arthritis, followed to a lesser extent by RF positivity and the presence of the shared epitope. Emery and colleagues have also examined this issue in their EAC and formulated guidelines for referral (Table 6-2). According to these guidelines, patients should be referred for an expedited rheumatologic evaluation if they have three or more swollen joints, involvement of MCP or MTP joints and morning stiffness for >30 minutes. The accuracy of an early RA diagnosis can be further enhanced by findings such as chronicity (>12 weeks of symptoms), elevated acute-phase reactants (ESR or CRP), or serologic abnormalities (RF or anti-CCP antibodies).

Visser analyzed data from 524 patients enrolled in the Leiden Early Arthritis Clinic and developed a prediction model for the development of persistent arthritis and erosive arthritis (assuming persistence is present). In this model, the best predictive factors for chronic arthritis were the presence of CCP antibodies, RF positivity, radiographic erosions, symptoms for >6 weeks but <6 months, polyarthritis and a positive MCP squeeze test (Table 6-3). Therefore, there are guidelines for high-yield referrals and simple measures to ensure the accuracy of an early RA diagnosis. Such guidelines help to clarify the clinician’s concern as to whether a patient’s articular complaints may be self-limiting or due to RA. An illustration of an MCP squeeze test is shown in Figure 6-5.

Enlarge 

Enlarge 

Enlarge 

The Challenge of Early RA

The challenge of early diagnosis and intervention will not be easily overcome. Patients may not recognize the importance of their symptoms or know where to acquire an accurate diagnosis and treatment. PCPs, rather than rheumatologists, are then likely to be the first-line care providers for those patients who seek medical attention. For them, the accuracy of an early RA diagnosis may be problematic because early manifestations are often nonspecific and nonprogressive and reliance on serologic diagnostic testing may be misleading since these tests have low sensitivity in patients with new-onset inflammatory arthritis. As discussed in Diagnostic Criteria, the 1987 classification criteria for RA have limited utility in early disease since they were established in patients with long-standing disease and thus require findings of chronic disease (e.g., nodules, radiographic erosions) to establish a diagnosis. The revised 2010 ACR/EULAR criteria place greater emphasis on lab tests (elevated ESR or CRP and seropositivity for RF or ACPA) and the number of joints, are more practical and more capable of an early RA diagnosis . One study in 205 patients found that the 2010 criteria were significantly better at detecting patients who would end up requiring DMARD therapy (62%, compared to 38% with the 1987 criteria; P <0.01). However, the converse was also true, with the 2010 criteria identifying more patients who ended up never requiring DMARD therapy because their disease resolved (8% vs 5% with the 1987 criteria; P = 0.01).

The benefits of early diagnosis are great but early diagnoses will not be easily achieved unless current clinical practice changes to meet the challenge. Although educational needs must be met, access to care and specialist consultation must also be facilitated by all practitioners, health care systems and insurers. Education of both PCPs and rheumatologists on the need for early diagnosis and benefits of early DMARD intervention have been shown to increase referrals and DMARD use in early disease.

The promise of earlier identification, referral and an accurate diagnosis can now be rewarded with highly effective monotherapy or combination DMARD or biologic therapies. Most practicing rheumatologists in North America are keenly interested in treating early RA, although most have not restructured their current practices to tackle this unmet need. Preclinical Rheumatoid Arthritis addresses how rheumatologists can better meet this challenge and ensure more effective collaboration with PCPs in the care of patients with early RA or undifferentiated RA.