`early rheumatoid arthritis
`
`T. Pincus1, Y. Yazici2, T. Sokka1,3, D. Aletaha4, J.S. Smolen4,5
`
`1Vanderbilt University, Nashville,
`Tennessee; 2Brooklyn Heights Arthritis
`Associates, Brooklyn, New York;
`3Jyväskylä Central Hospital, Jyväskylä,
`Finland; 4University of Vienna;
`5Krankenhaus Lainz, Vienna, Austria.
`Theodore Pincus, MD, Professor of Medi-
`cine; Yusuf Yazici, MD; Tuulikki Sokka,
`MD, PhD, Assistant Professor of Medicine;
`Daniel Aletaha, MD; Josef S. Smolen, MD,
`Professor of Medicine.
`Please address correspondence to:
`Theo-dore Pincus, MD, Professor of
`Medicine, Division of Rheumatology and
`Immunology, Vanderbilt University School
`of Medicine, 203 Oxford House, Box 5,
`Nashville, TN 37232-4500, USA.
`E-mail: t.pincus@vanderbilt.edu
`Supported in part by grants from Aventis,
`Amgen, Pfizer, the Jack C. Massey Founda-
`tion, the Academy of Finland and by NIH
`Grant HL 67964.
`Clin Exp Rheumatol 2003; 21 (Suppl. 31):
`S179-S185.
`© Copyright CLINICAL AND EXPERIMEN-
`TAL RHEUMATOLOGY 2003.
`
`Key words: Methotrexate, early
`rheumatoid arthritis.
`
`ABSTRACT
`The two major advances over the 1990s
`in the treatment of rheumatoid arthritis
`(RA) were a shift in strategy from a
`"pyramid", in which disease modifying
`anti-rheumatic drugs (DMARDs) were
`deferred for several years, to the early
`aggressive use of DMARDs and wide -
`spread acceptance of methotrexate as
`the DMARD with the most long-term
`effectiveness and safety. Methotrexate
`courses are continued far longer than
`those of any other DMARD, an excel -
`lent indicator of greater effectiveness
`and safe t y. In one recent seri e s ,
`m e t h o t rex ate was the fi rst DMARD
`used in more than 80% of patients with
`RA. Studies which document the supe -
`riority of combinations of methotrexate
`with biological agents to methotrexate
`monotherapy select for only a minority
`of contemporary patients with RA who
`have severe disease activity and incom -
`plete responses to methotrexate. In one
`locale, only 5% of patients met criteria
`for the A n t i - Tumor Necrosis Fa c t o r
`Trial in RA with Concomitant Therapy
`(ATTRACT) trial and only 30% met the
`c ri t e ria for the Early Rheumat o i d
`Arthritis (ERA) trial. In studies com -
`paring methotrexate directly with bio -
`l ogical age n t s , the biological age n t s
`have greater efficacy in patients with
`very severe disease, but the best results
`are seen in patients who take a combi -
`n ation of methotrex ate and biologi c
`agents. These data establish that
`methotrexate is the anchor drug and
`probably should be the first DMARD
`used in the majority of patients with RA
`at this time.
`
`Introduction
`The history of the treatment of rheuma-
`toid arthritis (RA) in the 20th century
`p resents a steady evolution of new
`agents and new ap p ro a ches. At the
`b eginning of the century,
`the only
`available drug therapy was aspirin (1).
`During the 1930s gold salts were intro-
`
`S-179
`
`duced by Forrestier and colleagues (2)
`and became the mainstay of therapy
`through the 1980s. Penicillamine was
`introduced in the 1970s (3), and anti-
`malarials gained widespread usage in
`the 1980s (4, 5). Sulfasalazine wa s
`actually developed in 1948, but did not
`reach widespread use until the 1980s
`(6).
`Each of these advances provided mean-
`ingful benefit to many patients in cop-
`ing more effe c t ive ly with their RA.
`However, despite the fact that rheuma-
`tologists spoke of secondary agents for
`the tre atment of RA as "re m i s s i o n
`inducing agents" (7), most pat i e n t s
`ex p e rienced progre s s ive disease, a n d
`RA was not adequately controlled in
`most patients (8-11). By contra s t , at
`this time control of RA appears to be a
`reasonable goal in most patients (12-
`14), comparable to the control of other
`chronic dysregulatory diseases such as
`hypertension and diabetes (15), albeit
`requiring ongoing therapy, as the etiol-
`ogy and treatment of the dysregulation
`remain unknown.
`Two major advances may account for
`the improved status of patients with RA
`over the last decade. The first involved
`a major shift in the strategies for patient
`c a re. Earlier ap p ro a ches had empha-
`sized deferring treatment with disease
`modifying
`anti-rheumatic
`dru g s
`(DMARDs) until disease had been pre-
`sent for a few years (16), explained in
`part by reports of population-based and
`clinical studies in the 1960s and 1970s
`that most people who met the criteria
`for RA appeared to have a good prog-
`nosis (17, 18). In add i t i o n , ava i l abl e
`DMARDs such as injectable gold salts
`and penicillamine had substantial toxi-
`cities and we re thought of as best
`avoided, wherever possible.
`During the 1980s it became apparent
`that most clinical patients with RA who
`were seen in rheumatology treatment
`settings had a progressive disease, in
`contrast to individuals seen in the early
`
`Page 1 of 7
`
`ANTARES Exhibit 1014
`
`
`
`Methotrexate as the "anchor drug" for early RA / T. Pincus et al.
`
`population and clinical studies. Impor-
`tant diffe rences we re
`re c og n i ze d
`between symptoms due to inflamma-
`tion, such as swollen joints or an ele-
`vated ery t h ro cyte sedimentation rat e
`( E S R ) , wh i ch we re reve rs i bl e, a n d
`symptoms due to joint damage which
`were cumulative and irreversible (19).
`S eve re long-term outcomes such as
`wo rk disab i l i t y,
`joint rep l a c e m e n t
`s u rge ry, and pre m at u re death we re
`common in many patients (20, 2 1 ) .
`Clinical trials ap p ro p ri at e ly incl u d e
`o n ly short - t e rm reve rs i ble measure s ,
`and suggested that good control of
`inflammation was seen over relatively
`short periods (22). However, long-term
`remission was unusual (11), and evi-
`dence of cumulative joint damage and
`poor long-term outcomes emerge d
`from longitudinal studies over 10 to 20
`years (8-11). A new approach to RA
`was proposed in the late 1980s involv-
`ing "remodeling the pyramid" (23-25)
`and thinking of RA as a "medical emer-
`gency" (26, 27), which requires early,
`aggressive intervention with a goal of
`remission, not mere improvement (12-
`14).
`The second major advance in the treat-
`ment of RA over the last two decades
`was a DMARD which was far more
`potent and safe than previously avail-
`able DMARDs – we e k ly low dose
`m e t h o t rex at e. Methotrex ate had been
`used in the 1960s in the treatment of
`inflammatory arthritides, but fell into
`disuse except in a few sites through the
`1970s and early 1980s, because it was
`felt to be too aggressive and toxic for
`the treatment of RA. A few pioneering
`r h e u m at o l ogists such as Hoff m e i s t e r
`(28, 29) and Scherbel (30, 31) treated
`patients who had RA with methotrexate
`during the 1960s and 1970s. This prac-
`tice led to clinical studies by Willkens
`(32), Weinstein (33), and others, and
`ultimately to a large multi-center clini-
`cal trial organized by Weinblatt (34),
`which clearly documented the efficacy
`and safety of methotrexate for the treat-
`ment of RA.
`Methotrexate is often included on lists
`of DMARDs as though it were one of a
`group of secondary agents for the treat-
`ment of RA. Howeve r, m e t h o t rex at e
`differs substantially from all available
`
`D M A R D s , s h owing gre ater effi c a cy
`and a high level of safe t y. New
`DMARDs such as cy cl o s p o rine (35,
`36) and leflunomide (37, 38), as well as
`the biologic agents – etanercept (39,
`40), infliximab (41), anakinra (42, 43)
`and adalimumab (44, 45) – represent
`major advances, providing mechanism-
`driven, targeted therapies for patients
`with RA. It is recognized that 20-30%
`of patients remain poorly contro l l e d
`with methotrexate and require further
`t h e rapy. Some patients have show n
`spectacular responses to anti-TNF
`agents. Nonetheless, methotrexate con-
`tinues to be the "anchor drug" for most
`patients with RA. It is generally the
`first drug used in the treatment of RA
`among the community of rheumatolo-
`gists in Nashville, Tennessee (46),
`although this is not the case in other
`contemporary rheumatology care set-
`tings, as discussed below.
`In this review, we summarize briefly
`the
`rationale
`for
`consideri n g
`m e t h o t rex ate as the anchor drug fo r
`RA, which is based on five phenomena:
`1. the excellent long-term effectiveness
`of methotrexate in most patients; 2. the
`l o n g - t e rm safety of methotrex ate in
`most patients; 3. the increasing use of
`methotrexate and its acceptance as the
`most effe c t ive DMARD by
`the
`rheumatology community; 4. recogni-
`tion that studies which document the
`superiority of biological agents or com-
`b i n ations of drugs with methotrex at e
`compared to methotrexate monothera-
`py select for a minority of contempo-
`rary patients with RA, who have severe
`disease
`activity
`and
`incomplete
`responses to methotrexate; and 5. evi-
`dence from early RA clinical trials that
`methotrexate is almost as effective as
`biological agents in patients with very
`severe RA.
`
`Long-term effectiveness of
`methotrexate in most patients
`Clinical observational studies and ran-
`domized controlled trials which estab-
`lished the efficacy of methotrexate in
`RA (29, 32-34, 47-50) were followed
`by careful long-term clinical observa-
`tional studies by We i n bl att and col-
`l e agues (51, 5 2 ) , K remer and col-
`leagues (53, 54), Sany and colleagues
`
`S-180
`
`( 5 5 , 56) and others. These rep o rt s
`cl e a rly established that methotrex at e
`was effective over long periods, with
`considerably lower toxicity than previ-
`ously available DMARDs.
`S eve ral long-term analyses of dat a
`from routine clinical care indicated that
`courses of methotrexate were contin-
`ued substantially longer than courses of
`other DMARDs, one of the best mea-
`sures of the long-term effectiveness of
`a DMARD. In 617 patients who had
`1,017 DMARD courses (57) and in 532
`patients in 7 US private practices (58),
`m o re
`than 50% of courses of
`m e t h o t rex ate we re continued over 5
`years or more, in contrast to fewer than
`20% of courses of injectable gold salts,
`p e n i c i l l a m i n e,
`hy d rox y ch l o ro q u i n e,
`and azathioprine. In 460 patients from
`7 private practices in Melbourne, Aus-
`tralia, 75.4% of patients were still tak-
`ing methotrexate after 6 years (59), and
`53% of patients were continuing at 12
`years (60).
`R ep o rts of improved mortality rat e s
`(61, 62) in patients with RA at this time
`compared to previous periods can be
`attributed in large part to methotrexate.
`Choi et al. recently reported that both
`m e t h o t rex ate and sulfasalazine we re
`m o re cost-effe c t ive than the new ly
`available treatment options of lefluno-
`mide and etanercept to achieve ACR 20
`responses and a weighted average of
`proportions achieving ACR 70, ACR 50
`and ACR 20 over a 6-month peri o d
`(63).
`
`Long-term safety of methotrexate
`M e t h o t rex ate has been one of most
`carefully studied DMARDs for adverse
`events associated with therapy. Th e
`experience at the Hospital for Special
`S u rge ry (64) indicated that only 94
`(3.4%) liver function tests out of a total
`of 2,791 performed in 182 RA patients
`were abnormal. One hundred fifty-two
`patients (83.5%) with 2007 evaluations
`had no ab n o rmal re s u l t s , c o m p a re d
`with 30 patients (16.5%) who had at
`least one abnormal liver function result
`in 784 tests. Twe n t y - t wo of the 30
`patients with at least one abnormality
`(73.3%) continued treatment despite an
`abnormality, without further evaluation
`or change in therapy, and subsequent
`
`Page 2 of 7
`
`
`
`Methotrexate as the "anchor drug" for early RA / T. Pincus et al.
`
`liver function assessments were within
`normal limits. The most common rea-
`son for discontinuation was inadequate
`response, and not side effects. These
`data were interpreted to suggest that
`guidelines developed by the ACR to
`monitor methotrex ate-taking pat i e n t s
`every 6 weeks may be in need of revi-
`sion, a suggestion supported in a sur-
`vey of U.S. rheumatologists (65).
`A review of 362 RA patients enrolled
`in an outpatient clinic at the Rheuma-
`tology Department of Vienna General
`Hospital indicated that liver enzyme
`ab n o rmalities
`in patients
`taking
`methotrexate virtually always occurred
`within the fi rst 4 months of therapy
`(66). These elevations did not lead to
`ch a n ges in therapy, and liver biopsy
`was not performed in any patients. The
`vast majority of laboratory abnormali-
`ties were fully reversible and no costly
`complications were seen. The data led
`to a suggestion that monitoring should
`be more frequent (every 2-4 weeks) in
`the first 4 months and then performed
`every 4-6 months, which was validated
`in another cohort of RA patients from
`another hospital in Vienna. It was cal-
`culated that a mean of 48-78% of costs
`could be saved if the proposal for less
`frequent monitoring was implemented
`(67).
`Methotrexate has a well-defined toxici-
`ty pro file and physicians monitor
`p atients for ga s t ro i n t e s t i n a l , h ep at i c,
`and pulmonary toxicity, bone marrow
`s u p p ression
`and
`stomatitis. A s
`methotrexate prescribing patterns have
`changed from initially being reserved
`for patients who had "climbed the RA
`treatment pyramid" to earlier in the dis-
`ease cours e, the toxicity pro file has
`i m p rove d. Patients are
`re l at ive ly
`healthier early in their disease and
`appear to be less vulnerable to ad verse
`events (12). In multiple cohort s ,
`methotrexate appears to have very few
`clinically significant side effects, possi-
`bly due in part to the routine use of
`folic acid supplementation (68).
`
`Increasing use of methotrexate and
`its acceptance as the most effective
`DMARD by rheumatologists
`When methotrexate was initially used
`by a large number of rheumatologists
`
`in the late 1980s, as noted above, it was
`generally begun after the patient had
`t ried (and failed) seve ral DMARDs,
`including injectable gold salts, penicil-
`l a m i n e, hy d rox y ch l o ro q u i n e, and (in
`Europe) sulfasalazine. For example, the
`report of 532 patients from 7 private
`US practices published in 1992 (58)
`i n d i c ated that methotrex ate was the
`fi rst DMARD used in 11.5% of
`p at i e n t s , c o m p a red to 38.9% start i n g
`with pare n t e ral go l d, 24.4% with
`hydroxychloroquine, 10.3% with peni-
`c i l l a m i n e, 16.9% with azat h i o p ri n e,
`and 1% with auranofin. Among 1,427
`p atients in Edmonton, Canada seen
`b e t ween 1985 and 1994, p a re n t e ra l
`gold was the most fre q u e n t ly pre-
`scribed initial DMARD from 1985 to
`1987, sulfasalazine from 1988 to 1990,
`and hy d rox y ch l o roquine after 1991,
`while methotrex ate was the initial
`DMARD in fewer than 5% of patients
`until 1994 (69). An analysis of 428
`patients with RA of less than one year's
`d u ration in Greece tre ated betwe e n
`1987-1995 indicated that the fi rs t
`DMARD was methotrexate in 27% of
`patients, hydroxychloroquine in 20%,
`penicillamine in 19%, cyclosporin in
`8 % , i n t ra - muscular gold in 7%, a n d
`other DMARDs in 21% of pat i e n t s
`(70).
`The early reluctance to use methotrex-
`ate as the initial DMARD may be based
`on at least three ex p l a n at i o n s : a )
`r h e u m at o l ogists had ex p e rience with
`more traditional DMARDs; b) a per-
`ception that the most potent drugs have
`the highest level of toxicities; c) con-
`c e rn about possible adve rse eve n t s
`involving hematologic and hepatic tox-
`icities, as well as a possible predisposi-
`tion to later malignancies. Howeve r,
`experience of more than 15 years has
`reinforced recognition of the long-term
`effectiveness, as well as safety, of low-
`dose weekly methotrexate, particularly
`the
`re c ognition
`that high dose
`m e t h o t rex ate and low-dose we e k ly
`methotrexate have very different toxici-
`ty profiles. Methotrexate has been used
`increasingly by rheumatologists as the
`initial DMARD in many settings, and
`attitudes about methotrex ate have
`changed considerably.
`Documentation of changes in clinical
`
`S-181
`
`practice may be seen in a number of
`reports. In a study of DMARD use in
`671 patients between 1975 and 1988,
`intra-muscular gold was taken by 100%
`of patients in 1975 versus fewer than
`2% in 1988, while methotrex ate use
`had changed from 0% up until 1980 to
`44% in 1988, although methotrex at e
`was ra re ly the fi rst DMARD wh e n
`these analyses were compiled in 1988
`(57). In Tromso, Norway methotrexate
`was used in 7% of patients between
`1979 and 1987 compared to 40% of
`p atients in 1988 through 1996 (71).
`Among the 593 patients monitored in
`Vienna, Austria, methotrexate was pre-
`scribed in fewer than 10% of patients
`prior to 1988 versus 38% of patients in
`1998, and was the initial DMARD in
`30% of patients in 1998 (67). In Fin-
`land, sulfasalazine was the most pre-
`s c ribed DMARD from 1995 thro u g h
`2000, but was overtaken by methotrex-
`ate in 2001 (72).
`The changing patterns of increased use
`of methotrexate in RA are reflected in
`s u rveys of rheumat o l ogists. A 1992
`survey in the United Kingdom indicat-
`ed that sulfasalazine was the most
`favored DMARD, as fewer than 10%
`of rheumatologists chose methotrexate
`as an initial DMARD (73). By contrast,
`a 2002 survey of 331 rheumatologists
`in the United Kingdom indicated that
`the first choice DMARD of 46.5% of
`r h e u m at o l ogists was methotrex at e
`c o m p a red to 43.5% who chose sul-
`fasalazine (74).
`Acceptance of methotrexate has been
`earlier and greater in North America
`than in Europe. A survey in the fall of
`1996 indicated that methotrexate was
`regarded as the first choice by 78.5% of
`214 United States rheumatologists and
`by 68.7% of Canadian rheumatologists
`(75). A survey of US rheumatologists
`in 1995 and 1999 indicated that 82%
`used a combination of methotrexate +
`hydroxychloroquine in 1995 compared
`to 96% in 1999, and 16% used combi-
`n ation DMARDs, wh i ch ge n e ra l ly
`i n cluded methotrex at e, in more than
`30% of patients in 1995 versus 46% in
`1999 (76).
`Reports of recent early RA databases in
`the United States indicate the more
`widespread use of methotrexate. In the
`
`Page 3 of 7
`
`
`
`Methotrexate as the "anchor drug" for early RA / T. Pincus et al.
`
`We s t e rn Consortium of Pra c t i c i n g
`R h e u m at o l ogists established betwe e n
`1993 and 1996 (77), methotrexate was
`used by 35.7% of patients at baseline
`and by 57.4% after two years. In the
`e a rly rheumatoid art h ritis tre at m e n t
`eva l u ation regi s t ry (ERATER) (46),
`m e t h o t rex ate was the fi rst DMARD
`used by 84.2% of pat i e n t s , and wa s
`used in 89% of patients seen in
`Nashville, Tennessee (none of whom
`were patients of any of the authors).
`Nonetheless, patients in this database
`f rom other sites showed diffe re n t
`t re n d s , i n cluding a group of pat i e n t s
`f rom Burl i n g t o n , M a s s a ch u s e t t s ,
`i n
`whom 37% we re
`tre ated with
`methotrexate and 40% with hydroxy-
`chloroquine as the first DMARD, and a
`group in Brooklyn, New York, in whom
`38% we re
`tre ated
`initially with
`methotrexate and 24% with hydroxy-
`chloroquine (78). In Europe, a slower
`acceptance of methotrexate can be seen
`in the Italian early arthritis database, in
`which methotrexate was used in 19%
`of patients with early rheumat o i d
`arthritis of less than 4 months' duration
`compared to 42% in those with arthritis
`of 4 months to 2 years duration (79).
`Methotrexate was used by only 4.6% of
`patients in the Norfolk arthritis register
`(NOAR) in the United Kingdom com-
`pared to sulfasalazine in 57% (80).
`One important further consideration is
`that weekly methotrexate therapy with
`doses of 10 mg per week or less may
`h ave limited effe c t ive n e s s , with sub-
`stantially lower retention rates than that
`seen for doses of 12.5 mg per week or
`m o re (81). Th e re fo re, m e t h o t rex at e
`doses of 15-25 mg should be given if
`tolerated. In many instances, parenteral
`administration of methotrexate results
`in both greater tolerability and greater
`efficacy. The most recent trials compar-
`ing the efficacy of methotrexate with
`that of biologicals employed high dose
`methotrexate therapy with rapid accel-
`eration of the dose.
`Taken together, although there remain
`disparities between beliefs and practice
`( 8 2 ) , these rep o rts indicate a tre n d
`t owa rds more widespread use of
`methotrexate by many rheumatologists
`in patients with early RA. The dat a
`suggest that many rheumatologists now
`
`rega rd methotrex ate as the pri m a ry
`"anchor drug" for treatment of RA.
`
`Studies documenting superiority of
`biological agents or drug combina-
`tions compared to MTX monothera-
`py select for only a minority of RA
`patients
`Over the last decade, a number of ran-
`domized controlled clinical trials have
`been published indicating greater effi-
`cacy for combinations of DMARDs or
`b i o l ogical agents with methotrex at e
`compared to methotrexate only, includ-
`ing cy cl o s p o rine (83),
`l e fl u n o m i d e
`(84), etanercept (39), infliximab (85),
`a d a l i mu m ab (45), and anakinra (43).
`H oweve r,
`these studies may have
`i n cluded only a small fraction of
`p atients with RA at the study sites,
`based on two important selection crite-
`ria which are sometimes neglected in
`the interpretation of the data.
`The first type of selection involves the
`"step up" or "add on" design of most
`studies, in which patients are eligible
`only if they respond incompletely to
`methotrexate (86). It would be expect-
`ed that patients who respond incom-
`pletely to any drug, whether an anti-
`hy p e rt e n s ive agent or even a non-
`steroidal anti-inflammatory drug, will
`respond with gre ater effi c a cy to the
`addition of a second drug versus the
`addition of a placebo. This is not to
`criticize the add-on clinical trial, which
`was developed at a time when the avail-
`able DMARDs we re not nearly as
`potent as the DMARDs available now,
`and the consensus was that combina-
`tion therapy offered no advantages over
`DMARD monotherapy (87,88). Efforts
`to document the potential efficacy of
`new DMARDs appeared to have been
`ove r whelmed by
`the substantially
`gre ater effi c a cy of methotrex ate than
`other available DMARDs. Therefore, it
`appeared appropriate to have patients
`obtain maximum
`effi c a cy
`fro m
`m e t h o t rex at e,
`and
`then
`analy ze
`whether an additional agent could pro-
`vide incremental effi c a cy. Howeve r,
`this pro c e d u re selected for pat i e n t s
`who we re poor
`re s p o n d e rs
`to
`methotrexate, who appear to represent
`a relative minority of patients.
`The "step up" or "add on" design is also
`
`S-182
`
`the most appropriate approach in test-
`ing a new agent in RA, when its effica-
`cy and toxicity are unknown, as it is
`ethical to offer the optimal ava i l abl e
`t h e rapy initially. Furt h e rm o re,
`t h e
`selection of partial responders is sensi-
`ble in order to exclude totally refracto-
`ry patients. However, there have been
`few analyses concerning how many
`patients seen in standard clinical care
`for RA who we re
`tre ated with
`methotrexate and other DMARDs may
`h ave been ineligi ble for inclusion in
`these clinical trials, because of a favor-
`able clinical status.
`The second type of selection involves
`inclusion criteria for contemporary RA
`clinical tri a l s , designed for pat i e n t s
`with the most severe RA. Most recent
`RA clinical trials continue to list inclu-
`sion criteria which were developed sev-
`e ral decades ago , s u ch as 6 tender
`joints, 6 swollen joints, an ESR ‡ 28
`mm/hour, and morning stiffness of ‡ 45
`minutes (89), although the clinical sta-
`tus of patients with RA appears to have
`improved substantially over this period
`(90,91). Two cohorts of patients seen in
`N a s h v i l l e, Tennessee we re rev i ewe d
`using a standard protocol for the evalu-
`ation of RA (SPERA) and were ana-
`lyzed to determine the proportion of
`patients who met 3 or 4 of these criteria
`(89). Cohort E (early) included 232
`patients with less than 3 years of symp-
`toms seen by 5 full-time private prac-
`tice rheumat o l ogists. Cohort L (lat e )
`included all 138 consecutive patients
`with RA (other than 14 who did not
`have a joint count recorded or who had
`t a ken infl i x i m ab or etanerc ep t ) , wh o
`had been under the care of one rheuma-
`t o l ogist (TP) at a we e k ly academic
`rheumatology clinic for a mean of 4.6
`years (range 0 – 19 years).
`Overall, 15.3% of Cohort L and 34.1%
`of Cohort E patients had 6 or more
`swollen and tender joints, as well as an
`ESR of 28 or more, or morning stiff-
`ness of 45 minutes or more (89). Only
`4.1% of Cohort L and no patients in
`Cohort E met ARA criteria for remis-
`sion. In analyses of specific clinical tri-
`als (92), among all 232 patients in
`Cohort E, 37 (16%) met inclusion crite-
`ria for the ERA clinical trial of
`m e t h o t rex ate ve rsus etanerc ept (40,
`
`Page 4 of 7
`
`
`
`Methotrexate as the "anchor drug" for early RA / T. Pincus et al.
`
`93). Among the 138 patients in cohort
`L who had a joint count recorded and
`we re not taking etanerc ept or infl i x-
`i m ab, o n ly 7 (5%) met the pri m a ry
`i n clusion cri t e ria for the A n t i - Tu m o r
`Necrosis Factor Trial in RA with Con-
`comitant Th e rapy (AT T R ACT) study
`(85, 94).
`It is recognized that the sponsors of
`these clinical trials deliberately sought
`patients with more severe disease, as
`should be the case with early clinical
`trials of new therapies. However, it is
`also recognized that most patients in
`the clinical cohorts from standard
`rheumatology care were ineligible for
`most contempora ry clinical tri a l s ,
`including the ERA trial and ATTRACT
`study. The majority of patients seen in
`the standard clinical cohorts were treat-
`ed with methotrexate, and had 1-5 ten-
`der or swollen joints and an ESR <28
`mm/hour (89) [up to 40% of patients
`have a normal ESR at their first visit
`(95)]. This observation suggests that
`methotrexate may be sufficient therapy
`for many, if not most, patients with RA,
`and/or that inclusion criteria for clini-
`cal trials might be broadened to be
`more generalizable.
`
`Efficacy of methotrexate in
`"head-to-head" comparisons with
`biological agents: Methotrexate is
`the "anchor" drug
`Several recent clinical trials compared
`methotrexate with TNF-blockers and/
`or a combination of TNF-blockers with
`m e t h o t rex at e. In these tri a l s , p at i e n t s
`received methotrexate at the start of the
`study rather than being partial respon-
`ders and continuing methotrexate.
`The ERA clinical trial to compare etan-
`e rc ept to methotrex ate in early RA
`patients with less than 3 years of dis-
`ease (40, 93) was discussed above and
`is presented in greater detail elsewhere
`in this supplement (96). The re s u l t s
`indicated superiority of etanercept over
`m e t h o t rex ate in ACR 20, 50 and 70
`responses at some time points, and in
`slowing the progression of total Sharp
`radiographic scores (40, 93). Many of
`these results are stat i s t i c a l ly signifi-
`c a n t , but diffe rences between etaner-
`cept and methotrexate are rather small,
`and their clinical significance is not
`
`e s t abl i s h e d. Furt h e rm o re, p atients in
`the ERA trial were selected for severity
`of RA, as fewer than 20% of 232
`patients with less than 3 years of dis-
`ease in one clinical setting met the
`inclusion criteria (92).
`Therefore, etanercept may be superior
`to methotrexate in patients who have
`severe clinical activity, i.e. 20-30% of
`patients. However, such patients are a
`minority of all patients with RA in sev-
`eral sites, including Norway (97). It is
`possible that most patients could do as
`well with methotrex at e, p o s s i bly in
`combination with hydroxychloroquine
`and/or sulfasalazine (98-100).
`In the ASPIRE trial wh i ch , l i ke the
`ERA trial, involved patients with < 3
`ye a rs disease durat i o n , m e t h o t rex at e
`was compared to a combination of
`methotrexate and infliximab. The com-
`bination was significantly superior to
`methotrexate monotherapy in all end-
`points – clinical, radiological and func-
`tional (101). In the TEMPO tri a l ,
`results of which were briefly presented
`at a EULAR Satellite symposium,
`patients with long-standing RA were
`t re ated with methotrex at e, e t a n e rc ep t
`or a combination of the two agents, and
`the combination was the most superior
`regimen.
`Thus, while monotherapy of biological
`agents may be only marginally superior
`to methotrex ate monotherapy (and
`might even be less so if methotrexate
`were combined with intermediate dose
`g l u c o c o rt i c o i d s ) , the combination of
`m e t h o t rex ate with TNF bl o cke rs
`appears to convey the maximal thera-
`peutic effects currently obtainable, at
`least in patients selected for hav i n g
`s eve re RA. In such an ap p ro a ch ,
`m e t h o t rex ate again serves as the
`"anchor" with which a biological agent
`can be combined for greater efficacy.
`Given that methotrexate may interfere
`p ri m a ri ly with IL-1 pat h ways (102),
`the combined bl o ckade of IL-1- and
`TNF-mediated pathologies may consti-
`tute one of several explanations for the
`s i g n i ficant effi c a cy observed by this
`type of combination therapy.
`
`Conclusion
`The data presented ab ove indicate a
`trend to increasing use of methotrexate
`
`S-183
`
`as the primary "anchor drug" for the
`treatment of RA, both as monotherapy
`and in combination therapies with
`other DMARDs or biologicals, includ-
`ing one report that methotrexate was
`the first DMARD used in more than
`80% of patients with early RA. These
`findings re flect the superior effi c a cy
`and safety of methotrexate compared to
`other DMARDs. Nonetheless, at least
`20-50% of patients do not continu e
`methotrexate for longer than 5 years,
`and therapy generally involves a life-
`time commitment, since the dysregula-
`tion that ch a ra c t e ri zes RA re m a i n s
`p o o rly understood and without any
`therapies. Therefore, there is clearly a
`need for additional DMARDs and bio-
`logical therapies to control RA in many
`patients at this time, although metho-
`trexate remains the anchor therapy for
`most patients.
`
`References
`1. RAGAN C: Rheumatoid arthritis: The natural
`history of the disease and its management.
`Bull NY Acad Med 1951; 27: 63-74.
`2. FORESTIER J: Rheumatoid arthritis and its
`treatment by gold salts. J Lab Clin Med 1935;
`20: 827-40.
`3. MU LT I C E N T R E TR I A L GRO U P: C o n t ro l l e d
`trial of D(-)penicillamine in severe rheuma -
`toid arthritis. Lancet 1973; 1: 275-80.
`4. MA K S Y M OWYCH W, RU S S E L L A S: A n t i-
`malarials in rheumatology:Efficacy and safe-
`ty. Semin Arthritis Rheum 1987; 16: 206-21.
`5. PAULUS HE: Antimalarial agents compared
`with or in combination with other disease-
`modifying antirheumatic drugs. Am J Med
`1988; 85 (Suppl. 4A): 45-52.
`6. HANNONEN P, MÖTTÖNEN T, H A KOLA M,
`O K A M: S u l fasalazine in early rheumat o i d
`arthritis. Arthritis Rheum 1993; 36: 1501-9.
`7. LIPSKY PE: Remission-inducing therapy in
`r h e u m atoid art h ritis. Am J Med 1983; 75
`(Suppl. 4B): 40-9.
`8. PINCUS T, CALLAHAN LF, SALE W G,
`B RO O K S A L , PAYNE LE, VAU G H N W K:
`Severe functional declines, work disability,
`and increased mortality in seve n t y - five
`r h e u m atoid art h ritis patients studied ove r
`nine ye a rs. A rt h ritis Rheum 1984; 27: 8 6 4 - 7 2 .
`9. S C OT T D L , G R I N D U L I S K A , S T RU T H E R S
`GR, COULTON BL, POPERT AJ, BACON PA:
`Progression of radiological changes in rheu-
`matoid arthritis. Ann Rheum Dis 1984; 43:8-
`17.
`10. RASKER JJ, COSH JA: The natural history of
`rheumatoid arthritis:A fifteen year follow-up
`study. The prognostic significance of features
`noted in the first year. Clin Rheumatol 1984;
`3: 11-20.
`11. WOLFE F, HAWLEY DJ: Remission in rheu-
`m atoid art h ritis. J Rheumatol 1985; 12: 2 4 5 - 5 2 .
`12. E M E RY P, SALMON M: E a rly rheumat o i d
`arthritis: Time to aim for remission ? Ann
`
`Page 5 of 7
`
`
`
`Methotrexate as the "anchor drug" for early RA / T. Pincus et al.
`
`Rheum Dis 1995; 54: 944-7.
`13. WEINBLATT ME: Rheumatoid arthritis: Treat
`now, not later ! (editorial). Ann Intern Med
`1996; 124: 773-4.
`14. PINCUS T, STEIN CM, WO L F E F: "No ev i-
`dence of disease" in rheumatoid art h ri t i s
`using methotrexate in combination with other
`drugs: A contemporary goal for rheumatol-
`ogy care ? Clin Exp Rheumatol 1997; 15:
`591-6.
`15. PINCUS T, GIBOFSKY A , W E I N B L ATT M E:
`Urgent care and tight control of rheumatoid
`arthritis as in diabetes and hypertension:bet-
`ter treatments but a shortage of rheumatolo-
`gists. Arthritis Rheum 2002; 46: 851-4.
`16. L I G H T F O OT RW JR: Tre atment of rheuma-
`toid arthritis. In MCCARTY DJ (Ed.): Arthri -
`tis and Allied Conditions. Philadelphia, Lea
`& Febiger, 1985: 668-76.
`17. M I K K E L S E N W M , DODGE H: A fo u r- ye a r
`follow-up of suspected rheumatoid arthritis:
`the Tecumseh, Michigan, community health
`study. Arthritis Rheum 1969; 12: 87-91.
`18. O'SULLIVAN JB, CATHCART ES: The preva-
`lence of rheumatoid arthritis: Follow-up eval-
`uation of the effect of criteria on rates in Sud-
`bury, Massachusetts. Ann Intern Med 1972;
`76: 573-7.
`19. PINCUS T, CALLAHAN LF: Prognostic mark-
`e rs of activity and damag