6odel.dt

The new england journal of medicine d r u g t h e r a p y
Therapeutic Strategies for Rheumatoid Arthritis heumatoid arthritis, a chronic, systemic, inflammatory From the Department of Internal Medicine,University of Nebraska Medical Center; autoimmune disease, has as its primary target the synovial tissues. When the disease is unchecked, it leads to substantial disability and premature death.
Center — both in Omaha. Address reprint It affects approximately 0.8 percent of adults worldwide,2 is more common in women requests to Dr. O’Dell at the Department (by a ratio of 3 to 1), and has an earlier onset in women, frequently beginning in the childbearing years. Recent advances in understanding the cytokine networks that are or at jrodell@unmc.edu.
responsible for the ongoing inflammatory response in rheumatoid arthritis4 have led to the successful use of therapies that target tumor necrosis factor a (TNF-a) and inter- Copyright 2004 Massachusetts Medical Society. leukin-1.5 (These therapies were discussed in a recent review in the Journal.6) During the past 10 years, improved understanding of the pathophysiology of rheu- matoid arthritis has led to several key changes in the approach to therapy. First, earlydiagnosis and treatment are important. Second, the use of disease-modifying anti-rheumatic drugs (DMARDs) in combination is highly effective. Third, the use of agentsthat target cytokines, such as TNF-a and interleukin-1, is an effective strategy. Andfourth, recognition is growing that the assessment of treatment outcomes should in-clude an analysis of important coexisting illnesses (particularly cardiovascular diseaseand osteoporosis). In this article, I will discuss the clinical application of these princi-ples, which has resulted in a marked improvement in clinical outcomes.7,8 e a r l y d i a g n o s i s a n d t r e a t m e n t Joint damage occurs early in the course of rheumatoid arthritis; 30 percent of patientshave radiographic evidence of bony erosions at the time of diagnosis, and this propor-tion increases to 60 percent by two years.9 Unfortunately, bony erosions and deformi-ties are largely irreversible. Initiation of therapy with DMARDs within three months afterthe diagnosis of rheumatoid arthritis is crucial; a delay of as little as three months in theintroduction of these medications results in substantially more radiographic damageat five years.10-12 Therefore, early diagnosis, although challenging,13 is critical.14,15 The diagnosis cannot be established by a single laboratory test or procedure but is aided by the use of seven diagnostic criteria16 that favor clinical factors and, therefore,depend on the clinician’s asking insightful questions and recognizing the often-subtleearly physical findings. The diagnostic criteria are the presence of morning stiffness,arthritis of three or more joint areas, arthritis of the hand joints, symmetric arthritis,rheumatoid nodules, elevated levels of serum rheumatoid factor, and radiographicchanges. Many other syndromes, including self-limiting viral conditions lasting severalweeks, mimic rheumatoid arthritis. Therefore, the first four criteria must be presentfor a minimum of six weeks before a diagnosis of rheumatoid arthritis can be made.
This approach, however, leads to diagnostic uncertainty that may delay appropriatetherapy for months or years. Serum antibodies have been detected that may help define Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine subgroups of patients.17 Antibodies to cyclic citrul- nsaids
linated peptides (CCPs) appear to have a high spec- NSAIDs are particularly helpful during the first few
ificity (90 to 98 percent) and thus may prove useful weeks in which a patient has symptoms, because
in early diagnosis, even though the sensitivity of the the drugs provide partial relief of pain and stiffness
test for them is approximately 50 to 65 percent at until a definitive diagnosis of rheumatoid arthritis
presentation.18,19 Interestingly, these antibodies can be established. NSAIDs have not been shown to
may appear in the serum years before the onset of slow the progression of the disease; therefore, in
clinical disease.20 In any event, many patients pre- long-term care, NSAIDs should be used together
sent initially to primary care physicians, so these with DMARDs.21 Although both these classes of
providers need to recognize patients with potential medications are well tolerated for short periods,
inflammatory arthritis and be aware of the impor- long-term administration may result in gastrointes-
tance of referral to a specialist within the first three tinal ulcer, perforation, and hemorrhage. Every year
months after the appearance of symptoms.
1.5 percent of patients with rheumatoid arthritisare hospitalized with gastrointestinal problems.23The risk of these complications increases with older g e n e r a l t h e r a p e u t i c p r i n c i p l e s age, corticosteroid use, and a history of peptic ulcer Guidelines concerning therapy for rheumatoid ar- disease.
thritis have been published recently by the American Recently, cyclooxygenase-2 (COX-2) inhibitors, College of Rheumatology (Fig. 1).21 No treatment which decrease the incidence of gastric and duode-cures rheumatoid arthritis; therefore, the therapeu- nal ulcers by approximately 50 percent as comparedtic goals are a remission of symptoms involving the with traditional NSAIDs, have been introduced.24-26joints, a return of full function, and the maintenance To a similar degree, the addition of proton-pumpof remission with DMARD therapy. A useful inter- inhibitors to therapy with NSAIDs also decreasesmediate goal is to have all patients evaluated by a the incidence of bleeding ulcers associated with tra-rheumatologist within three months after the onset ditional NSAIDs.27 The efficacy of the COX-2 inhib-of symptoms, so that essentially all patients will be itors is no better than that of the older and less ex-receiving DMARDs by the time they have had symp- pensive NSAIDs.24 Both traditional NSAIDs andtoms for three months.
COX-2 inhibitors have been associated with in- To evaluate the success of interventions, inves- creased fluid retention, exacerbation of hyperten- tigators have used a number of clinical measures. sion, and impairment of renal function in suscep-
They include the number of joints that are tender tible patients.24,28 Thrombotic events have been
and swollen, markers of inflammation (including reported in patients who are taking COX-2 inhibi-
the erythrocyte sedimentation rate and C-reactive tors29-31 and may occur more frequently than with
protein level), and patients’ responses to questions traditional NSAIDs.29,30
about their pain, their global assessment of disease
activity, and their physical function. The American corticosteroids
College of Rheumatology criteria for improvement Corticosteroids are potent suppressors of the in-
can be used by clinicians to quantify patients’ im- flammatory response in rheumatoid arthritis and
provement after treatment (Table 1).22 Most clini- in many other diseases; unfortunately, their dose-
cal studies have required a benchmark of 20 percent dependent side effects are familiar to all clinicians.
improvement in these criteria, a result that is known
as ACR 20; now that better treatments are being
initiated earlier in the course of the disease, 50 per-
Figure 1. Management of Rheumatoid Arthritis.
cent improvement (ACR 50) is becoming a more Since many patients with rheumatoid arthritis present initially to primary care physicians, these providers need to suspect this diagnosis early and promptly refer the patient to a rheumatologist. Optimal care of patients with rheumatoid arthritis requires close collaboration between rheumatolgists and primary care physicians. Medications that are used to treat rheumatoid arthri- DMARD denotes disease-modifying antirheumatic drug, NSAID nonsteroidal antiinflammatory drug, and MTX tis are divided into three main classes: nonsteroidal methotrexate. Adapted from American College of Rheu- antiinflammatory drugs (NSAIDs), corticosteroids, and DMARDs (both synthetic and biologic).
Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. Establish diagnosis earlyDocument baseline disease activity are Physician
Start DMARD therapy within 3 moConsider NSAIDConsider local or low-dose systemic Primary C
Rheumatologist
Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 1. American College of Rheumatology Preliminary Definition
Table 2. Keys to Optimizing the Outcome of Treatment.*
of 20 Percent Improvement in Rheumatoid Arthritis (ACR 20).
Measure of Disease Activity
Requirement
Start DMARD therapy as early as possible (within three Strive for remission (no joint symptoms) in all patients Use corticosteroids as a bridge to effective DMARD Patient’s global assessment of disease activity Prednisone at doses >10 mg/day is rarely indicated Physician’s global assessment of disease activity Patient’s assessment of physical function Avoid using corticosteroids without DMARDs Minimize duration and dose by tapering to the low- Always consider prophylaxis to avert osteoporosis Controversy continues about when, if, and how these compounds should be used to treat rheu- Facilitate communication between primary care physi- matoid arthritis.32,33 Studies conducted more re-cently34-36 have corroborated findings from older * DMARD denotes disease-modifying antirheumatic drug.
trials37,38 by clearly establishing that corticosteroidsdecrease the progression of rheumatoid arthritis asdetected radiographically. Corticosteroids in low ications that retard or halt the progression ofdoses (e.g., ≤10 mg of prednisone per day) are used disease. Disease modification is most convincinglyto treat 30 to 60 percent of patients.21,32 The major- demonstrated by the ability of the medications toity of patients who were enrolled in recent pivotal decrease radiographic progression. A meta-analysisclinical trials were receiving corticosteroids at base- of blinded clinical trials has suggested that theline.39-44 Table 2 presents some useful guidelines relative efficacy of methotrexate, sulfasalazine, in-for corticosteroid use in patients with rheumatoid tramuscular gold, and penicillamine is similar.52arthritis.
Antimalarial drugs (e.g., chloroquine and hydroxy- Predictable side effects of corticosteroid drugs chloroquine) are less effective. Penicillamine, be- include thinning of the skin, cataracts, osteopo- cause of concern about its toxicity, and oral gold,rosis, hypertension, and hyperlipidemia.45,46 The because of its marginal efficacy,52 are rarely usedlatter three conditions may be preventable with ag- today.53 gressive management of osteoporosis and cardio- Since observational trials have clearly identified vascular risk factors. Essentially, all patients taking methotrexate as the synthetic DMARD that is mostcorticosteroids should receive supplemental calci- likely to induce a long-term response,54,55 it is mostum (1 to 1.5 g per day) and vitamin D (800 IU per often selected for initial therapy.53 It has demon-day).47 Bisphosphonates appear to be very effective strated efficacy and durability,56 a long-term trackin reducing vertebral fractures in patients taking record of acceptable toxicity, and low cost. An im-corticosteroids47 (offering a reported 70 percent re- portant observational study has shown that patientsduction in incidence) and should be prescribed for with rheumatoid arthritis who have been treatedpatients who have low bone density (e.g., a T score with methotrexate have significantly lower mortalityof ¡2 or lower). Recent evidence suggests that in- (odds ratio for death, 0.4) than patients who haveflammation plays a major role in atherosclero- not been treated with methotrexate.57 Methotrexatesis48-51; the ability of corticosteroids to decrease has become so much the standard of care that mostinflammation rapidly may offset some of the poten- of the recent pivotal trials of its use in patients withtially detrimental effects of the drugs on the vascu- established disease (Fig. 2) have listed active diseaselar endothelium.51 despite methotrexate therapy as an inclusion crite-rion.39,42-44 Concomitant administration of folic s y n t h e t i c d m a r d s
acid (1 to 3 mg per day) or folinic acid (2.5 to 5 mg Optimal management of rheumatoid arthritis re- given 12 to 24 hours after methotrexate) significant-quires rapid and sustained suppression of inflam- ly decreases many toxic effects without a measur-mation with DMARDs, which are defined as med- able decrease in efficacy and has improved the tol- Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. erability of methotrexate,61,62 a finding that haspermitted clinicians to administer 20 to 30 mg of methotrexate per week when necessary. Most data suggest that methotrexate is most effective at a doseof 17.5 to 30 mg per week63,64 but that oral absorp- tion may be highly variable.65,66 Therefore, if oral methotrexate produces a suboptimal response, a trial of subcutaneous or intramuscular methotrex- Leflunomide, a new synthetic DMARD, has an ef- ACR 20 Response (%)
ficacy similar to that of sulfasalazine or moderate- dose methotrexate.67,68 Hydroxychloroquine, con- sidered the least potent but best tolerated of the DMARDs,52 is the DMARD that is most commonly combined with methotrexate.53 Sulfasalazine was the first DMARD that was developed specifically to treat rheumatoid arthritis69 and has an efficacy sim-ilar to that of methotrexate.52 Sulfasalazine is most Figure 2. Responses to Drug Therapy in Seven Studies Involving Patients
commonly used in the United States as part of com- Receiving Methotrexate.
bination DMARD therapy.53 Intramuscular gold is All the patients in these seven studies were already receiving methotrexate, to which the other medication was added. Responses were measured in terms more likely both to produce a remission and to re- of an improvement of at least 20 percent in symptoms, as defined by the sult in toxic effects as compared with methotrex- American College of Rheumatology (ACR 20). Numbers above the bars are ate.70 Its cumbersome administration and toxicity Historically, concern about the toxicity of DMARDs has delayed their use in treating rheuma-toid arthritis. It is now accepted that the conse- acute pneumonitis is rare with methotrexate (51quences of delaying therapy far outweigh the pos- cases have been reported worldwide) but may besible toxic effects for the majority of patients.21,71 life-threatening.82 If pneumonitis is suspected onSafe administration of DMARDs requires critical the basis of clinical findings or a chest radiograph,and careful monitoring (Table 3).72 Detailed mon- methotrexate should be promptly discontinued anditoring guidelines have been published to help avert not reintroduced. As the best tolerated of all thedamage to the liver,73 which was a major concern DMARDs, hydroxychloroquine83 — if it is used atwhen use of the drug for rheumatoid arthritis be- doses below 6.5 mg per kilogram of lean body masscame popular in the mid-1980s.74-77 Since then per day — requires only yearly visits to the ophthal-only one case of cirrhosis has been reported in pa- mologist to prevent the rare occurrence of retinaltients whose physicians were following the guide- toxic effects.84lines for monitoring.78 These guidelines include Four double-blind, controlled trials have now measuring serum albumin and aminotransferase shown that minocycline is effective in treating rheu-levels every four to eight weeks. Doses of metho- matoid arthritis.85-88 In patients with long-stand-trexate should be decreased when aminotransfer- ing disease, a small but statistically significantase levels are elevated above the upper limit of benefit was observed.85,86 Minocycline that wasnormal, and treatment should be stopped if the ele- used as initial therapy in patients who tested posi-vation persists. Obtaining complete blood counts tive for rheumatoid factor was superior to placeboand measuring serum creatinine are also recom- (response rate, 65 percent, as compared with 13mended,72 since a decrease in renal function may percent for placebo87) and superior to hydroxy-precipitate toxic effects in a patient in previously chloroquine (60 percent vs. 33 percent88) whenstable condition who is taking methotrexate.79 measuring ACR 50. The mechanism by which mi- Both methotrexate and leflunomide have a sub- nocycline works is incompletely understood but stantial potential for teratogenesis,80,81 so women probably involves immunomodulation,89 suppres-of childbearing potential who require these medi- sion of matrix metalloproteinases,90 and suppres-cations should be using reliable birth control. Sub- sion of nonspecific infections that would otherwise Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 3. Guidelines for Monitoring the Treatment of Rheumatoid Arthritis.*
Potential
Baseline
System Review
Toxic Effects
Evaluation
or Examination
Laboratory Tests
Comments
CBC, creatinine, urine Rash, mouth ulcers, Creatinine every 2 weeks Poor long-term continua- Infections; symptoms None unless patient Infections; symptoms None unless patient Infections; symptoms None unless patient * DMARD denotes disease-modifying antirheumatic drug, CBC complete blood count,G6PD glucose-6-phosphate dehydrogenase, ALT alanine aminotransferase, and CHF congestive heart failure. The CBC includes a platelet count in all cases listed. Measurement of aspartate amino-transferase may be substituted for measurement of ALT.
stimulate inflammatory cytokine production.88 Re- and one that inhibits the action of interleukin-1versible hyperpigmentation is seen in up to 30 per- (anakinra) are now available to treat rheumatoidcent of patients who are receiving long-term mino- arthritis.6 Other agents are being tested; some havecycline therapy.
targeted cytokines in ways that are similar to thosepreviously mentioned. There is also renewed inter- b i o l o g i c d m a r d s
est in products that target the activation of T cells Three biologic products that inhibit the actions of or B cells,91,92 including anti-CTLA4Ig93,94 andTNF-a (infliximab, etanercept, and adalimumab) anti-CD20 (rituximab).95 (A report on a study of Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. rituximab appears elsewhere in this issue of the triple therapy (methotrexate, sulfasalazine, and hy-Journal.96) droxychloroquine) over sulfasalazine alone,99 dou-ble therapy (methotrexate and sulfasalazine or c o m b i n a t i o n t h e r a p y w i t h d m a r ds
methotrexate and hydroxychloroquine), and mono- Ten years ago, the use of combinations of DMARDs therapy.96 However, no trials have compared initialwas rare; now at least one third of patients with combination therapy with a rapid step-up to com-rheumatoid arthritis who are treated by rheuma- binations only in patients with active disease de-tologists in the United States are receiving com- spite monotherapy.
bination therapy.53 Trials that have compared In the much-studied group of patients with ac- combinations of DMARDs head-to-head with meth- tive disease despite taking methotrexate, a numberotrexate40,97 and trials that have shown the addi- of trials with similar designs have been completed.
tional benefit of adding drugs to methotrexate for In these trials, patients continue taking methotrex-patients who have active disease despite metho- ate plus either the active treatment or placebo. Fig-trexate39,42-44,58-60 (Fig. 2) have fueled the rise of ure 2 details the ACR 20 responses seen in sevencombination DMARD therapy.
trials involving patients with these characteristics.
One of the first randomized studies that direct- This summary of the data is not intended to imply ly compared combination DMARD therapy with that a specific level of response in one trial can bemethotrexate was a two-year, double-blind trial in directly compared with that in another trial. Thesewhich patients were assigned to three groups: seven effective therapies have not yet been com-those who took methotrexate alone (at 17.5 mg per pared head to head.
week), those who took a combination of sulfasal-azine (at 1 g per day) and hydroxychloroquine (at 400 mg per day), and those who took all three med-ications.40 At two years, the end point of 50 percent Establishing a diagnosis as early as possible andimprovement in composite symptoms of arthritis then starting DMARD therapy is the foundation forwas reached by 77 percent of patients who were successful treatment of patients with rheumatoidtreated with all three drugs but by only 33 percent arthritis. Many questions remain, including the fol-of patients who were treated with methotrexate lowing: Which drug should be used first? Can drugalone. Patients who received combination therapy toxicity or a patient’s response be predicted by phar-did not have more side effects than those who re- macogenomics? Should drug combinations be usedceived methotrexate alone. In another study, the initially? Should we try to induce rapid disease sup-triple combination of methotrexate (at 17.5 mg per pression with corticosteroids or TNF inhibitors?week), sulfasalazine (at 2 g per day), and hydroxy- Are there biologic markers that should be used tochloroquine (at 400 mg per day) was superior to monitor outcome in place of or in addition to stan-either the combination of methotrexate and sulfa- dard assessment?salazine or the combination of methotrexate and Until these questions are answered, most rheu- matologists select methotrexate as the initial ther- In patients who have early disease, three criti- apy for most patients.53 The characteristics and per- cal trials have all shown that initial combination sonal choices of patients influence this decision.21therapy is superior to therapy with a single Methotrexate should not be used in patients whoDMARD.96,98,99 The Combinatietherapie Bij Reu- have underlying liver or renal disease, who con-matoide Artritis (COBRA)98 trial compared sulfa- sume alcohol, who plan to become pregnant in thesalazine alone with the combination of sulfasala- near future, or who do not want to undergo regularzine, low-dose methotrexate (which was stopped laboratory monitoring. Whether to start a course ofat 40 weeks), and prednisolone (which was given low-dose corticosteroids initially along with theinitially at 60 mg per day but tapered off by 28 chosen DMARD is controversial; many cliniciansweeks). Patients in the combination group had a start treatment with prednisone at 5 to 7.5 mgmore rapid response to treatment, fewer withdraw- per day as bridge therapy until the slower-actingals from the study because of toxicity, and most im- DMARDs have a chance to work. Once the DMARDportant, less radiographic evidence of progression begins working, corticosteroids should be taperedat five years.100 Other trials involving patients with (Table 2).
early disease have demonstrated the superiority of Methotrexate is started at a dose of 7.5 to 15 mg, Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine given orally once weekly. If patients continue to shown in Figure 2, with the exception of poor long-have active disease (as indicated by swollen and term tolerability of combining methotrexate andtender joints), as most do, the dose should be in- cyclosporine.103 Until there are studies that com-creased in 5-mg increments each month or two to pare the relative efficacy of these effective therapies,20 to 30 mg per week. If patients continue to have the most economical initial choice104 for the patientactive disease, consideration should be given to who has active disease despite taking methotrex-switching to subcutaneous administration.21,65,66 ate is the addition of sulfasalazine, hydroxychloro-If active disease persists despite optimal metho- quine, or both. Data from the United Kingdomtrexate therapy, other DMARDs should be added.21 have shown that if conventional DMARDs are opti- Compelling data from the COBRA trial98,100 — mized and used in combination, control of the dis- in which patients with active early rheumatoid ar- ease can be achieved in more than half of patientsthritis were given either sulfasalazine alone or the who would otherwise be candidates for TNF inhib-combination of sulfasalazine, methotrexate, and itors.105 If active disease (manifested by swolleneither a high or a low dose of oral prednisolone for and tender joints) persists after three months of28 weeks — have raised the question of whether these DMARD combinations, leflunomide or a TNFtherapy to suppress the disease rapidly should be inhibitor should be added to methotrexate.
given immediately after the diagnosis. The rapid If DMARD therapy is started within three months suppression of rheumatoid arthritis by corticoste- after the onset of symptoms and escalated with theroids has been recognized since the middle of the goal of achieving remission, the majority of patients20th century.101 The Early Rheumatoid Arthritis will have their disease well controlled within a year(ERA) trial102 is an important study that compared while taking conventional single or combinationmethotrexate (at a dose that escalated to 20 mg per DMARD therapy.106 If inflammatory disease is in-week) with etanercept in patients with disease diag- adequately controlled, therapy with TNF inhibitorsnosed within the preceding three years. Both treat- should be started.
ments were very effective in controlling the diseaseat one year, but etanercept (administered at 25 mg subcutaneously twice a week) was more effective inrapidly suppressing disease activity. In both the ERA The long-term prognosis for patients with rheuma-and COBRA trials, markers of inflammation (in- toid arthritis depends not only on how well theircluding the erythrocyte sedimentation rate and the joint disease is treated but also on how well their co-C-reactive protein level) were dramatically reduced existing illnesses are addressed.107,108 The threeafter two weeks of therapy. If, in fact, it is important coexisting conditions that have the greatest effectto control disease in days, rather than in weeks or on morbidity and mortality in rheumatoid arthritismonths, then corticosteroids and TNF inhibitors are infection (particularly pulmonary infection), os-(both of which appear to be capable of stopping ac- teoporosis, and cardiovascular disease.
tive disease) should be examined in trials to test the Rheumatoid arthritis is associated with approx- concept of induction therapy (i.e., medication that imately a doubling of the risk of infection, as com-is administered initially and then withdrawn).
pared with the risk in age-matched controls,109 andthe degree of increase in this risk correlates withthe severity of the disease.110 Although some studies t h e r a p y f o r e s t a b l i s h e d suggest that corticosteroids may increase the risk of infection,111 controversy exists about whether If patients continue to have active disease after two such an increase is due to the use of corticosteroidsto three months of methotrexate at a dose of 20 to itself or to the fact that patients who are at higher30 mg per week, or if they cannot tolerate higher risk are more likely to use corticosteroids. Whetherdoses of methotrexate despite folate replacement, the new TNF inhibitors increase the risk of infec-the current standard practice is to add another tion is a matter for debate since the drugs have beenDMARD to methotrexate21; all the trials included associated with a change in the spectrum of infec-in Figure 2 address this patient population.
tions112 — specifically, increased tuberculosis, his- To date, clinically significant differences have not toplasmosis, and listeria.6 emerged with regard to toxicity for the treatments The clinician who is caring for patients with Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. rheumatoid arthritis should be aware of the risk of infection. All patients should have yearly influenzavaccinations and should receive the pneumococcal Caring for patients with rheumatoid arthritis posesvaccine at appropriate intervals. Since patients may significant challenges. Three issues deserve specialhave a better immunologic response to vaccination mention: the lack of an accurate method for makingbefore taking methotrexate,113 it seems prudent to an early diagnosis, which permits early treatment;vaccinate before starting DMARD treatment, when inadequate predictors of the differential responsepossible. Live vaccines should be avoided in patients to available therapy, as well as too few studies com-who are receiving immunosuppressive medica- paring effective therapies; and the enormous cost oftions. When considering TNF inhibitors, clinicians new therapies, which has made them unavailableshould recommend that all patients be tested for to many patients. Since musculoskeletal disease ac-prior exposure to tuberculosis. Both clinicians and counts for one quarter of all visits to the offices ofpatients with rheumatoid arthritis should be vigi- primary care physicians,120 formal training in rheu-lant with regard to avoiding infections and treating matic disease needs to be included in the educationthem early and aggressively. Stopping or withdraw- of all primary care physicians. Preliminary studiesing drug treatment during infections is critical.112 suggest that genetics might be useful in predicting The advent of diagnostic tools that are accurate responses to DMARDs.121 Until these research ef- and easy to use and the availability of effective ther- forts come to fruition, studies that directly compareapies for osteoporosis have been tremendous ad- current therapies are urgently needed. Many of ourvances in the treatment of rheumatoid arthritis.114 new therapies are expensive, with initial costs thatThe incidence of osteoporosis is doubled in patients may exceed $1,500 per month. However, these up-with rheumatoid arthritis,114,115 and baseline bone- front costs may be justified in the long term by sav-density studies should be performed in all patients, ings that result from an improved quality of lifeparticularly those who will receive corticosteroids. and enhanced productivity.
If osteoporosis is present, bisphosphonate therapy, The treatment of rheumatoid arthritis has im- which is reported to decrease the risk of fracture by proved dramatically in the past decade,7,8 thanks70 percent despite the coadministration of cortico- to early diagnosis and the availability of DMARDs.
steroids,47 should be used.
Physicians can now have the goal of eradicating ac- Cardiovascular disease accounts for most of tive disease and aggressively intervening to address the excess mortality associated with rheumatoid coexisting illnesses. Remission in patients who aredisease.49 The newer concepts of the pathogenesis receiving therapy is now a realistic goal. Recent re-of atherosclerosis suggest that inflammation is a ports that only a small minority of patients (i.e.,key factor in causing vascular endothelial dam- 5 percent) who are being treated by rheumatolo-age.49,50,116 It has been hypothesized that the sys- gists have disease that is active enough to qualifytemic inflammation that characterizes rheumatoid them for current clinical trials are a testament toarthritis may play a key role in the excess athero- the success of these therapies.107sclerosis seen in patients with this disease.49,50 Risk Supported by the Albert G. and Bernice F. Hansen Charitable factors for atherosclerosis should be aggressively Foundation.
Dr. O’Dell reports having received consulting fees from Amgen, sought and addressed. In particular, smoking cessa- Abbott, Centocor, Wyeth, and Bristol-Myers Squibb and lecturetion may be fruitful, since smoking has been asso- fees from Thompson, Armand Scott, and Maritz.
ciated with increased severity of arthritis.117 Current I am indebted to Renee Crosby and Lucie Case for their assistance in the preparation of the manuscript; to Drs. Arthur Weaver, Lynell trials of statin therapy in patients with rheumatoid Klassen, Iain McInnes, Gerald Moore, Ted Mikuls, Joel Bessmer,arthritis may address the risk of cardiovascular and David O’Dell for their support and thoughtful and candid reviewdisease, since statins should decrease both athero- of the manuscript; and to Donna O’Grady, Colleen Buescher, and r e f e r e n c e s
matoid arthritis. Rheum Dis Clin North Am results from the Norfolk Arthritis Register.
rheumatoid arthritis: results after 20 years.
CR, Scott DG, Silman AJ. The incidence of ways and joint inflammation in rheumatoid rheumatoid arthritis in the United Kingdom: arthritis. N Engl J Med 2001;344:907-16.
Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine thritis from undifferentiated polyarthritis in patients with early arthritis. J Rheumatol joint destruction in rheumatoid arthritis.
20. Nielen MM, van Schaardenburg D, Ree-
35. Hickling P, Jacoby RK, Kirwan JR. Joint
sink HW, et al. Specific autoantibodies pre- destruction after glucocorticoids are with- cede the symptoms of rheumatoid arthritis: a study of serial measurements in blood do- term functional disability in rheumatoid ar- nors. Arthritis Rheum 2004;50:380-6.
36. van Everdingen AA, Jacobs JWG, Sie-
thritis from 1977 to 1998: a longitudinal 21. American College of Rheumatology Sub-
study of 3035 patients. Am J Med 2003;115: dose prednisone therapy for patients with early active rheumatoid arthritis: clinical rheumatoid arthritis: 2002 update. Arthritis efficacy, disease-modifying properties, and O’Fallon WM, Matteson EL. Declining use side effects: a randomized, double-blind, of orthopedic surgery in patients with rheu- 22. Felson DT, Anderson JJ, Boers M, et al.
placebo-controlled clinical trial. Ann Intern matoid arthritis? Results of a long-term, rheumatoid arthritis. Arthritis Rheum 1995; 37. A comparison of prednisolone with as-
pirin or other analgesics in the treatment of 23. Singh G. Recent considerations in non-
rheumatoid arthritis. Ann Rheum Dis 1959; patients with early rheumatoid arthritis. Br J steroidal anti-inflammatory drug gastropa- thy. Am J Med 1998;105:Suppl 1B:31S-38S.
38. A comparison of prednisolone with as-
10. Lard LR, Visser H, Speyer I, et al. Early
24. FitzGerald GA, Patrono C. The coxibs,
pirin or other analgesics in the treatment of versus delayed treatment in patients with re- selective inhibitors of cyclooxygenase-2.
rheumatoid arthritis: a second report by the cent-onset rheumatoid arthritis: compari- son of two cohorts who received different 25. Bombardier C, Laine L, Reicin A, et al.
Council and Nuffield Foundation on clinical Comparison of upper gastrointestinal toxic- trials of cortisone, ACTH, and other ther- ity of rofecoxib and naproxen in patients ateutic measures in chronic rheumatic dis- 11. Egsmose C, Lund B, Borg G, et al. Pa-
39. Tugwell P, Pincus T, Yocum D, et al.
from early 2nd line therapy: 5-year followup 26. Silverstein FE, Faich G, Goldstein JL, et
Combination therapy with cyclosporine and of a prospective double blind placebo con- al. Gastrointestinal toxicity with celecoxib vs methotrexate in severe rheumatoid arthritis.
trolled study. J Rheumatol 1995;22:2208-13.
12. Tsakonas E, Fitzgerald AA, Fitzcharles
osteoarthritis and rheumatoid arthritis: the 40. O’Dell JR, Haire CE, Erikson N, et al.
MA, et al. Consequences of delayed therapy CLASS study: a randomized controlled trial.
with second-line agents in rheumatoid ar- methotrexate alone, sulfasalazine and hy- thritis: a 3 year followup on the Hydroxy- 27. Chan FKL, Hung LCT, Suen BY, et al.
droxychloroquine, or a combination of all chloroquine in Early Rheumatoid Arthritis Celecoxib versus diclofenac and omeprazole three medications. N Engl J Med 1996;334: (HERA) study. J Rheumatol 2000;27:623-9.
in reducing the risk of recurrent ulcer bleed- 13. Koopman WJ, ed. Arthritis and allied
ing in patients with arthritis. N Engl J Med 41. Maini RN, Breedveld FC, Kalden JR, et
conditions: a textbook of rheumatology.
al. Therapeutic efficacy of multiple intrave- 14th ed. Vol. 1. Philadelphia: Lippincott 28. Whelton A, Maurath CJ, Verburg KM,
nous infusions of anti-tumor necrosis factor Williams & Wilkins, 2001:1153-74.
Geis GS. Renal safety and tolerability of 14. Quinn MA, Conaghan PG, Emery P. The
celecoxib, a novel cyclooxygenase-2 inhibi- therapeutic approach of early intervention tor. Am J Ther 2000;7:159-75. [Erratum, Am for rheumatoid arthritis: what is the evi- 29. Food and Drug Administration. Arthri-
42. Weinblatt ME, Kremer JM, Bankhurst
tis Advisory Committee open panel meeting AD, et al. A trial of etanercept, a recombi- 15. O’Dell JR. Treating rheumatoid arthritis
2/8/2001: review of VIGOR, cardiovascular nant tumor necrosis factor receptor:Fc fu- early: a window of opportunity. Arthritis safety review of rofecoxib. (Accessed May 5, sion protein, in patients with rheumatoid 2004, at http://www.fda.gov/ohrms/dockets/ arthritis receiving methotrexate. N Engl J 16. Arnett FC, Edworthy SM, Bloch DA, et
30. Strand V, Hochberg MC. The risk of car-
43. Lipsky PE, van der Heijde DMFM, St
1987 revised criteria for the classification of diovascular thrombotic events with selective Clair EW, et al. Infliximab and methotrexate rheumatoid arthritis. Arthritis Rheum 1988; cyclooxygenase-2 inhibitors. Arthritis Rheum in the treatment of rheumatoid arthritis.
17. Goldbach-Mansky R, Lee J, McCoy A, et
31. Crofford LJ, Oates JC, McCune WJ, et al.
44. Cohen S, Hurd E, Cush J, et al. Treat-
al. Rheumatoid arthritis associated autoan- Thrombosis in patients with connective tis- ment of rheumatoid arthritis with anakinra, a tibodies in patients with synovitis of recent sue diseases treated with specific cyclooxy- genase-2 inhibitors: a report of four cases.
18. Schellekens GA, de Jong BA, van den
trexate: results of a twenty-four-week, multi- Hoogen FH, van de Putte LB, van Venrooij WJ.
32. Moreland LW, O’Dell JR. Glucocorti-
center, randomized, double-blind, placebo- Citrulline is an essential constituent of anti- coids and rheumatoid arthritis: back to the controlled trial. Arthritis Rheum 2002;46: future? Arthritis Rheum 2002;46:2553-63.
toid arthritis-specific autoantibodies. J Clin 33. Saag KG, Criswell LA, Sems KM, Nettle-
45. Saag KG, Koehnke R, Caldwell JR, et al.
Low dose long-term corticosteroid therapy 19. Jansen ALMA, van der Horst-Bruinsma
roids in rheumatoid arthritis: a meta-analy- in rheumatoid arthritis: an analysis of serious IE, van Schaardenburg D, van de Stadt RJ, de sis of their moderate-term effectiveness.
adverse events. Am J Med 1994;96:115-23.
46. McDougall R, Sibley J, Haga M, Russell
toid factor and antibodies to cyclic citrulli- 34. Kirwan JR, Arthritis and Rheumatism
nated peptide differentiate rheumatoid ar- arthritis receiving prednisone compared to Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. et al. Concomitant leflunomide therapy in tis: suggested guidelines for monitoring liv- patients with active rheumatoid arthritis de- er toxicity. Arthritis Rheum 1994;37:316-28.
47. American College of Rheumatology Ad
spite stable doses of methotrexate: a ran- 74. Weinblatt ME, Coblyn JS, Fox DA, et al.
domized, double-blind, placebo-controlled Efficacy of low-dose methotrexate in rheu- trial. Ann Intern Med 2002;137:726-33.
prevention and treatment of glucocorticoid- 61. Morgan SL, Baggott JE, Vaughn WH, et
induced osteoporosis: 2001 update. Arthri- al. Supplementation with folic acid during 75. Thompson RN, Watts C, Edelman J, Es-
methotrexate therapy for rheumatoid arthri- daile J, Russell AS. A controlled two-centre 48. Libby P, Ridker PM, Maseri A. Inflamma-
tis: a double-blind, placebo-controlled trial.
trial of parenteral methotrexate therapy for tion and atherosclerosis. Circulation 2002; refractory rheumatoid arthritis. J Rheumatol 62. Van Ede AE, Laan RF, Rood MJ, et al.
49. Van Doornum S, McColl G, Wicks IP.
Effect of folic or folinic acid supplementa- 76. Williams HJ, Willkens RF, Samuelson
Accelerated atherosclerosis: an extraarticu- tion on the toxicity and efficacy of metho- CO Jr, et al. Comparison of low-dose oral lar feature of rheumatoid arthritis? Arthritis trexate in rheumatoid arthritis: a forty-eight pulse methotrexate and placebo in the treat- ment of rheumatoid arthritis: a controlled 50. Del Rincon I, Williams K, Stern MP,
blind, placebo-controlled study. Arthritis clinical trial. Arthritis Rheum 1985;28:721- Freeman GL, O’Leary DH, Escalante A. As- sociation between carotid atherosclerosis 63. Furst DE, Koehnke R, Burmeister LF,
77. Andersen PA, West SG, O’Dell JR, Via
and markers of inflammation in rheumatoid Kohler J, Cargill I. Increasing methotrexate arthritis patients and healthy subjects. Ar- effect with increasing dose in the treatment methotrexate in rheumatoid arthritis: clini- of resistant rheumatoid arthritis. J Rheuma- 51. Hallgren R, Berne C. Glucose intoler-
ance in patients with chronic inflammatory 64. Genovese M, Keystone EC, Tesser JRP,
diseases is normalized by glucocorticoids.
78. Kremer JM. Not yet time to change the
oral methotrexate (MTX) in early RA is need- 52. Felson DT, Anderson JJ, Meenan RF.
liver toxicity: they have served us well.
The comparative efficacy and toxicity of sec- ond-line drugs in rheumatoid arthritis: re- 65. Herman RA, Veng-Pedersen P, Hoffman
79. O’Dell JR. Methotrexate use in rheuma-
sults of two metaanalyses. Arthritis Rheum of low-dose methotrexate in rheumatoid ar- 53. Mikuls TR, O'Dell J. The changing face
thritis patients. J Pharm Sci 1989;78:165-71.
80. Janssen NM, Genta MS. The effects of
of rheumatoid arthritis therapy: results of 66. Brooks PJ, Spruill WJ, Parish RC, Birch-
ry medications on fertility, pregnancy, and lactation. Arch Intern Med 2000;160:610-9.
54. Pincus T, Marcum SB, Callahan LF.
81. Nelson JL, Ostensen M. Pregnancy and
Long-term drug therapy for rheumatoid ar- rheumatoid arthritis. Arthritis Rheum 1990; rheumatoid arthritis. Rheum Dis Clin North thritis in seven rheumatology private prac- tices. II. Second line drugs and prednisone.
67. Smolen JS, Kalden JR, Scott DL, et al. Ef-
82. Kremer JM, Alarcon GS, Weinblatt ME,
ficacy and safety of leflunomide compared et al. Clinical, laboratory, radiographic, and 55. Wolfe F, Hawley DJ, Cathey MA. Termi-
with placebo and sulphasalazine in active histopathologic features of methotrexate- nation of a slow acting antirheumatic ther- rheumatoid arthritis: a double-blind, ran- associated lung injury in patients with rheu- apy in rheumatoid arthritis: a 14-year pro- matoid arthritis: a multicenter study with literature review. Arthritis Rheum 1997;40: starts. J Rheumatol 1990;17:994-1002.
68. Strand V, Cohen S, Schiff M, et al. Treat-
56. Ortendahl M, Holmes T, Schettler JD,
83. Felson DT, Anderson JJ, Meenan RF. Use
Fries JF. The methotrexate therapeutic re- of short-term efficacy/toxicity tradeoffs to sponse in rheumatoid arthritis. J Rheumatol select second-line drugs in rheumatoid ar- thritis: a metaanalysis of published clinical 57. Choi HK, Hernán MA, Seeger JD, Rob-
69. Svartz N. Salazopyrin, a new sulfanil-
trials. Arthritis Rheum 1992;35:1117-25.
ins JM, Wolfe F. Methotrexate and mortality 84. Marmor MF, Carr RE, Easterbrook M,
in patients with rheumatoid arthritis: a pro- spective study. Lancet 2002;359:1173-7.
70. Rau R, Herborn G, Menninger H, Blech-
58. Weinblatt ME, Keystone EC, Furst DE, et
necrosis factor alpha monoclonal antibody, in the treatment of early erosive rheumatoid for the treatment of rheumatoid arthritis in arthritis: 12 month data of a double-blind 85. Kloppenburg M, Breedveld FC, Terwiel
patients taking concomitant methotrexate: parallel study of 174 patients. Br J Rheuma- JP, Mallee C, Dijkmans BA. Minocycline in active rheumatoid arthritis: a double-blind, 48:35-45. [Erratum, Arthritis Rheum 2003; 71. Pincus T, Callahan LF. The ‘side effects’
placebo-controlled trial. Arthritis Rheum of rheumatoid arthritis: joint destruction, 59. O’Dell JR, Leff R, Paulsen G, et al. Treat-
disability and early mortality. Br J Rheuma- 86. Tilley BC, Alarcon GS, Heyse SP, et al.
Minocycline in rheumatoid arthritis: a 48- 72. American College of Rheumatology Ad
week, double-blind, placebo-controlled trial.
trexate and sulfasalazine, or a combination Hoc Committee on Clinical Guidelines.
of the three medications: results of a two- Guidelines for monitoring drug therapy in 87. O’Dell JR, Haire CE, Palmer W, et al.
rheumatoid arthritis. Arthritis Rheum 1996; Treatment of early rheumatoid arthritis with controlled trial. Arthritis Rheum 2002;46: minocycline or placebo: results of a ran- 73. Kremer JM, Alarcon GS, Lightfoot RW
domized, double-blind, placebo-controlled 60. Kremer JM, Genovese MC, Cannon GW,
Jr, et al. Methotrexate for rheumatoid arthri- trial. Arthritis Rheum 1997;40:842-8.
Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved. 88. O’Dell JR, Blakely KW, Mallek JA, et al.
109. Doran MF, Crowson CS, Pond GR,
Treatment of early seropositive rheumatoid sulphasalazine with sulphasalazine alone in O’Fallon WM, Gabriel SE. Frequency of in- arthritis: a two-year, double-blind compari- early rheumatoid arthritis. Lancet 1997;350: fection in patients with rheumatoid arthritis compared with controls: a population-based quine. Arthritis Rheum 2001;44:2235-41.
99. Möttönen T, Hannonen P, Leirisalo-
study. Arthritis Rheum 2002;46:2287-93.
89. Ritchlin CT, Haas-Smith SA, Schwarz
110. Wolfe F, Mitchell DM, Sibley JT, et al.
therapy with single-drug therapy in early The mortality of rheumatoid arthritis. Ar- doxycycline upregulates IL-10 production in rheumatoid arthritis: a randomised trial.
human synoviocytes, mononuclear cells and 111. Doran MF, Crowson CS, Pond GR,
synovial explants. Arthritis Rheum 2000;43: 100. Landewe RBM, Boers M, Verhoeven
O’Fallon WM, Gabriel SE. Predictors of in- fection in rheumatoid arthritis. Arthritis 90. Golub LM, Evans RT, McNamara TF,
patients with early rheumatoid arthritis: long-term structural benefits of a brief inter- 112. Ellerin T, Rubin RH, Weinblatt ME. In-
crobial tetracycline inhibits gingival matrix vention. Arthritis Rheum 2002;46:347-56.
fections and anti-tumor necrosis factor a 101. Hench PS, Kendall EC, Slocumb CH,
therapy. Arthritis Rheum 2003;48:3013-22.
phyromonas gingivalis-induced periodonti- Polley HF. The effect of a hormone on the 113. O’Dell JR. Gilg J, Palmer W, Haire C,
tis in rats. Ann N Y Acad Sci 1994;732:96- adrenal cortex (17-hydroxy-11-dehydrocor- ticosterone: compound E) and of pituitary in rheumatoid arthritis: decreased response 91. Panayi GS, Corrigall VM, Pitzalis C.
while taking methotrexate. J Clin Rheuma- Pathogenesis of rheumatoid arthritis: the toid arthritis: preliminary report. Proc Staff role of T cells and other beasts. Rheum Dis 114. Haugeberg G, Orstavik RE, Uhlig T,
102. Bathon JM, Martin RW, Fleischmann
Falch JA, Halse JI, Kvien TK. Bone loss in pa- 92. Zhang Z, Bridges SL Jr. Pathogenesis of
RM, et al. A comparison of etanercept and tients with rheumatoid arthritis: results rheumatoid arthritis: role of B lymphocytes.
methotrexate in patients with early rheuma- from a population-based cohort of 366 pa- Rheum Dis Clin North Am 2001;27:335-53.
toid arthritis. N Engl J Med 2000;343:1586- tients followed up for two years. Arthritis 93. Kremer JM, Westhovens R, Leon M, et
al. Treatment of rheumatoid arthritis by se- 115. Ridker PM, Rifai N, Rose L, Buring JE,
lective inhibition of T-cell activation with fu- 103. Yocum DE, Stein M, Pincus T. Long-
Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol lev- els in the prediction of first cardiovascular 94. Weinblatt M, Schiff M, Goldman M,
methotrexate (MTX) in the treatment of ac- events. N Engl J Med 2002;347:1557-65.
Kremer J, Breazna A, Becker J-C. A pilot, tive rheumatoid arthritis (RA): analysis of 116. Michel BA, Bloch DA, Fries JF. Predic-
tors of fractures in early rheumatoid arthri- placebo controlled study of a co-stimulation 104. Choi HK, Seeger JD, Kuntz KM. A cost
117. Mattey DL, Hutchinson D, Dawes PT,
effectiveness analysis of treatment options et al. Smoking and disease severity in rheu- rheumatoid arthritis. Arthritis Rheum 2002; for methotrexate-naive rheumatoid arthri- matoid arthritis: association with polymor- phism at the glutathione S-transferase M1 95. Edwards JCW, Szczepanski L, Szechin-
105. Bingham SJ, Buch MH, Lindsay S, Ten-
locus. Arthritis Rheum 2002;46:640-6.
ski J, et al. Efficacy and safety of rituximab, a nant A, Emery P. The impact of escalating 118. Blake GJ, Ridker PM. Are statins anti-
B-cell targeted chimeric monoclonal anti- therapy in rheumatoid arthritis patients re- inflammatory? Curr Control Trials Cardio- ferred for anti-tumour necrosis factor-alpha trial in patients with rheumatoid arthritis.
119. Leung BP, Sattar N, Crilly A, et al.
A novel anti-inflammatory role for simva- 106. Sokka T, Pincus T. Eligibility of pa-
statin in inflammatory arthritis. J Immunol 96. Edwards JCW, Szczepa´
tients in routine care for major clinical trials ski J, et al. Efficacy of B-cell–targeted therapy of anti-tumor necrosis factor alpha agents in 120. Spitzer WO, Harth M, Goldsmith CH,
with rituximab in patients with rheumatoid rheumatoid arthritis. Arthritis Rheum 2003; et al. The arthritic complaint in primary care: arthritis. N Engl J Med 2004;350:2572-81.
prevalence, related disability, and costs.
97. Calgüneri M. Pay S, Caliskaner Z, et al.
107. Mikuls TR, Saag KG. Comorbidity in
rheumatoid arthritis. Rheum Dis Clin North 121. O’Dell JR, Nepom BS, Haire C, et al.
for the treatment of patients with rheuma- toid arthritis. Clin Exp Rheumatol 1999;17: 108. Gabriel SE, Crowson CS, O’Fallon
predicting response to specific treatments.
WM. Mortality in rheumatoid arthritis: have 98. Boers M, Verhoeven AC, Markusse HM,
we made an impact in 4 decades? J Rheuma- Copyright 2004 Massachusetts Medical Society. Downloaded from www.nejm.org at UNIVERSITY OF GRONINGEN on March 21, 2006 . Copyright 2004 Massachusetts Medical Society. All rights reserved.

Source: http://studiekringnoord.nl/Syllabus/syllabus2006/bijdragen/7/Therapeutic%20strategies%20for%20RA%20O'Dell.pdf