Management of therapy-resistant rheumatoid arthritis

BaillieÁre's Clinical Rheumatology Vol. 13, No. 4, pp. 737±752, 1999

12 Management of therapy-resistant rheumatoid arthritis Eric-Jan J. A. Kroot

MD

Leo B. A. van de Putte Piet L. C. M. van Riel

MD, PhD

MD, PhD

Department of Rheumatology, University Hospital Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands

During the last two decades, newly introduced therapeutic strategies have resulted in satisfactory modi®cation of the disease course in the majority of the patients with rheumatoid arthritis (RA). Nevertheless, a de®nite number of RA patients remain therapy-resistant, and for this group more aggressive treatment may be required for preventing permanent disability and progressive joint damage necessitating surgical procedures. Therefore, management of therapy-resistant RA is one of the major challenges in modern rheumatology. RA patients who have not responded to conventional disease-modifying antirheumatic drug (DMARD) therapy are de®ned as refractory RA patients. However, a uniform description or de®nition for `refractory' RA does not appear to be available. In this article we will deal with, and discuss, the term `refractory RA' based on a MEDLINE database search using this term, currently available therapeutic options, data on therapy-resistant RA patients from an inception cohort of RA patients attending the Nijmegen University Hospital, management of extra-articular manifestations and future management strategies. Key words: rheumatoid arthritis; refractory; treatment; extra-articular manifestations.

PRACTICAL MANAGEMENT OF DIFFICULT CLINICAL SITUATIONS BASED ON AVAILABLE EVIDENCE AND EXPERIENCE Refractory rheumatoid arthritis: urgent need for a validated de®nition! The term `refractory' RA is commonly used to indicate treatment resistant RA. However, standard textbooks do not provide a concise description of the term `refractory'. Therefore, we searched the MEDLINE database from February 1992 to October 1998 using the MeSH headings: `arthritis', `rheumatoid' and `refractory'. A total of 137 articles were found, but only full length articles in English were included in the analysis. Not included were case reports, editorials, observational and nonrandomized studies. In this way 40 studies were found using the term `refractory'. All articles were screened by one author (EJK). Data extracted from these studies included baseline patient characteristics (number of patients, number of previously used drugs, 1521-6942/99/040737+16 $12.00/00

c 1999 Harcourt Publishers Ltd. *

738 E.-J. J. A. Kroot et al

disease duration, disease activity at inclusion), study-design, previous drug treatment, and study medication. Surprisingly, the characterisation of these refractory patients di€ered widely between these studies. Most of these studies were clinical trials. A great di€erence occurred in the number of DMARDs used before inclusion of the patients. The minimal number of DMARDs used before study entry ranged from one to more than ®ve, and in most studies patients had been using at least 3 or 4 DMARDs. In the majority of the studies patients were treated with `conventional DMARDs', but data on dosage regimens and treatment duration were frequently not mentioned. With regard to previous drug treatment, methotrexate appeared to be the most commonly prescribed DMARD. Di€erences between these studies were not only noticed in terms of the minimum number of DMARDs used, but also in terms of disease duration and disease activity. As can be seen from Table 1, disease duration data indicated that most patients had longlasting disease, which correlates with the high mean number of DMARDs used (4±5). Disease activity entry criteria were used in almost all studies, but di€ered both in type and degree. In 75% of the studies, inclusion criteria were de®ned as having at least six tender joints, six swollen joints, morning sti€ness lasting 45 minutes and an erythrocyte sedimentation rate (ESR) of at least 28 mm/hour. In addition, in two studies the term refractory was also used to indicate mono-articular active disease (one active joint in the presence of otherwise well-controlled disease). Analysing the 40 therapeutic studies using the term `refractory', it can be concluded that this term is not well de®ned at the moment and as these studies show, di€er widely in terms of de®nitions of previous treatment, disease duration and disease activity. Most studies, in particular clinical trials, have used the term `refractory' for inclusion of patients. As therapeutic response depends on de®nition or composition of the selected group of patients, the di€erent studies are hardly comparable, because of the di€ering de®nitions. For instance, response in long-standing disease is only achieved when real active disease is present at the time these patients are included in clinical trials. Long-standing disease may also respond di€erently from early disease, but data con®rming this are currently not available. In order to compare these clinical trials, European League against rheumatism (EULAR)1 and American College of Rheumatology (ACR)2 core-set variables are used. For further characterisation of the term `refractory' RA, speci®c disease characteristics should be used. As can be concluded from Table 1, these characteristics should minimally include disease duration and activity, and treatment characteristics such as the number of previous DMARDs, dosage and treatment duration. In the future, current management strategies will probably be further re®ned, so the de®nition of `refractory' may well be changed. In conclusion, it has become apparent that there is an urgent need for an exact de®nition of the term `refractory', in which a minimal set of disease characteristics are included. `Refractory' RA in an inception cohort of recently diagnosed RA patients As there is no generally accepted de®nition of the term refractory, we have tried to characterise this term in an inception cohort of RA patients with recently diagnosed RA (symptoms 51 year at study entry). Characterisation of this term was primarily based on 40 recent studies using this term. Starting in 1985, all consecutive patients attending the Nijmegen University Hospital with recent onset RA (51 year), according to the American Rheumatism Association criteria (ARA, 1987), were asked to participate in a long-term prospective study. Up to the end of 1997, 350 patients who

Therapy-resistance management 739

had not previously received second-line antirheumatic drugs entered the study. Since 1985 all patients have been followed up in a standardized way. The patients were treated by di€erent rheumatologists, and were seen every three months by specially trained research nurses. Quantitative clinical and laboratory data were collected, and the use of DMARDs and non-steroidal anti-in¯ammatory drugs (NSAIDs) in respect of starting date, dose, stopping date, reason for discontinuation and side-e€ects was recorded. As treatment strategies change frequently, for example upper limits of DMARD dosages, an exact characterisation of the term refractory is dicult to obtain. Nevertheless, we studied the extent of therapy-resistant RA patients in this unique inception cohort, using prescription patterns at the authors' department. Sulphasalazine appeared to be the most frequently prescribed initial DMARD, whereas methotrexate was the most frequently prescribed DMARD of second choice. As a meta-analysis has shown that sulphasalazine and methotrexate are the best choices with regard to toxicity and ecacy3, minimum usages for these most commonly used DMARDs have been imposed: (1) sulphasalazine had to be used for at least 3 months, in a minimal daily dosage of 2000 mg; (2) in addition, methotrexate had to be used for at least 3 months as well, with a minimal weekly dosage of 15 mg before stopping the drug; (3) as well as these two DMARDs, a third DMARD, being one of the other conventional DMARDs (gold, azathioprine, d-penicillamine, antimalarials or cyclophosphamide) must have been used. No minimum dosages or treatment durations have yet been de®ned for these drugs. Summarizing, patients were considered to have therapy-resistant RA, if they had been treated with at least three di€erent DMARDs, including both methotrexate, sulphasalazine and a third DMARD. In this way, analysing `refractory' RA, of the total study group of 350 patients, 210 patients, with a mean disease duration of 9 years, were included. During the period of observation 61 patients used 3 or more DMARDs. However, of this group 34 patients were still using (and responding to) one of these drugs at the end-point of our observations, so could not be de®ned as having refractory RA. Therefore, only 27 patients ful®lled the criteria, primarily based on 40 previous studies, using the term `refractory'. The mean therapy duration for methotrexate was 20 months (range 3±53) and for sulphasalazine 21 months (range 3±78). Lack of ecacy appeared to be the most important reason for discontinuing sulphasalazine (23 patients). The reasons for discontinuation of methotrexate therapy appeared to be more varied: toxicity (8 patients); lack of ecacy (5 patients); pregnancy and hospital admission (5 patients) were the most frequently reported reasons. In 7 patients, methotrexate was prescribed twice, in all cases in combination with sulphasalazine, but was in these cases discontinued again because of lack of ecacy or toxicity. Sulphasalazine was prescribed twice in only 2 patients, in both cases in combination with methotrexate, and was in both cases discontinued because of lack of ecacy. Surprisingly, out of the 27 patients ful®lling the criteria, 12 patients were treated with only three di€erent DMARDs, whereas they were not using any DMARD at the follow up end-point. In conclusion, therapy-resistant RA was observed in a small number of recent onset RA patients (+10%) for up to 12 years (mean 9 years) of follow up. This small number may even be further reduced when the term `refractory' is further de®ned. However, our ®rst de®nition of the term `refractory' was primarily based on 40 recent studies using this term and on the prescription patterns of DMARDs at our department. The preliminary results of analysing this term in an inception cohort of RA patients have shown that more speci®c criteria are needed for further characterisation of this

Keysser Moreland46 Schirmer47 Isaacs48 Moreland49 Moreland50 Kovarik51 Maksymowych52 Moreland53 Schrohenloher54 Lacki55 Isaacs56 Sala57 Kanik58 Blanco59 Isaacs60 Malaise61 Weinblatt62 Moreland63 Matteson64 Moreland65 Maksymowych66

45

First author

22 23 101 102 9 12 13 14 15 16 17 18 11 19 20 21 24 25 26 27 28 29

open db pc open open db open open open db db pc open open open pilot retro open open open db pc open db pc open

Design

108 18 5 9 180 25 20 4 16 22 28 9 28 20 43 41 59 40 64 30 45 19

Patients (n) no def 10 13, 2 12 80%45y 11 no def 13 8, 5 11 5, 8 10 6, 7 no def 8 11, 3 13, 4 14, 5 8, 7 8 9, 5 13, 2

no def 1 3 2 1 1 1 4 1 2 4 1 1 1 1 3 4 1 2 1 2 2

Disease Previous (years) drugs (n) no def MIT, HCQ, SASP, penic, gold conv. DMARDs all pts used Auro HCQ, SASP, MTX, AZA, penic HCQ, SASP, gold penic 1 DMARD MTX DMARDs DMARDs SASP, gold, penic, MTX SASP, penic, gold, 60% MTX MTX or HCQ 90% MTX, 10% SASP Single DMARD therapy DMARDs DMARDs 90% MTX MTX for at least 23 weeks DMARDs MTX, SASP, AZA, HCQ, gold, penic DMARDs

Drugs used

no def ARAcrit t410, s410, b445 t410, s410, b430 t410, s410, b428 t49, s46, b428 t46, s43, b428 t415, s415 t49, s45 t49, s46 severe t410, s410, b428 ARAcrit t48, s42, HAQ45 ARAcrit t46, s43, ms445 t46, s43, ms460, b428 t46, s43, ms445, b429 t49, s46, ms445, b428 t46, s43, ms445, b428 t49, s46, ms445, b428 t49, s46, b410

Disease activity at entry

Table 1. Characteristics of 40 studies concerning patients with refractory rheumatoid arthritis.

44, 5 years 4 weeks 8 weeks 3 months 3 months 6 months 6 months 3 months 2 months 4 weeks 1 year 7 days 6 months 18 weeks 6 months 10 days 6/12 months 8 months 3 months 6 months 3/4 months 3 months

Study duration

CYC anti-CD4 cladribine anti-CD4 anti-TNF MTX/anti-CD4 diclo/CsA IVIg anti-TNF IL-2 cycloph/pred iv CAMPATH-1H CsA, MTX or HCQ IVIg or albumin MTX and AZA CAMPATH-1H CsA CAMPATH-1H MTX and anti-CD4 CAMPATH-1H IL-2 pentoxifylline

Active drug

740 E.-J. J. A. Kroot et al

37 38 39 42 43 44 40 46 47 48 49 50 51

Flippo72 Benenson73 Polisson74 Moreland75 Moreland76 Weusten77 Sewell83 Woodworth78 Racadot79 Isaacs80 Tumiati81 Altman82 Shiroky7

open open rc open retro open open open retro open db pc open open open open open rc

open open open letter open

31 42 148 25 25 50 19 45 17 8 10 11 8

40 32 25 16 11 11 7 13, 5 11 11 11 10, 5 no def no def 13 4, 8 7, 6 17

ARAcrit 15, 2 10, 1 15, 5 12, 4 5 5 no def 1 1 2 4 2 1 2 1 1 3

1 2 2 4 3

Table 1 (continued).

DMARDs (100% MTX), cortico's DMARDs (10% MTX) and cortico's DMARDs gold, SASP, HCQ, penic gold, SASP, HCQ, penic DMARDs MTX and gold DMARDs (100% MTX) Steroids DMARDs (100% SASP, 63% MTX) DMARDs (60% MTX, 0% SASP) HCQ or MTX 100% MTX, 100% gold, 37% SASP

gold or MTX for 6 months DMARDs MTX HCQ, AZA, gold, penic DMARDs, 100% MTX, 80% SASP 45 DMARDs incl. MTX functional class 2 t49, s46, ms445, b428 t49, s46, ms445, b428 t49, s46, ms445, b428 active t46, s43, ms445, b428 no def steroid-refr t410, s410, ms445, b430 t46, s43, ms445, b430 t46, s43, ms445, b440 t410, s410, ms445, b430

t49, s49, ms445, b428 t49, s46, ms445, b428 t49, s46, ms445, b428 t410, s410, b428 t420, s46, ms430, b428 1 year 6 months 18 weeks 6 months 6 months 3, 8 years 4 weeks 1 year 1 year 8 months 6 months 20 weeks 6 months

12 months 3 months 30 months 4 years fup 24 weeks

CsA, MTX or gold Mab to ICAM-1 anti-CD4 and MTX MTX and prednison PROSORBA, protein a CsA prospidine or MTX MTX or placebo anti-CD4 anti-CD4 corticopulsetherapy IL-2 IL-2 anti-CD4 CAMPATH-1H IVIg CsA and or nsaid MTX and leucovorin

open, open study/trial; db, double-blind; pc, placebo controlled; retro, retrospective; rc, randomized controlled no def, not de®ned; t, tender-joint count; s, swollen-joint count; ms, morning sti€ness; b, erythrocyte sedimentation rate; ARAcrit, ARA 87 criteria for RA; MTX, methotrexate; SASP, sulphasalazine; HCQ, hydroxychloroquine; penic, d-penicillamine; CsA, cyclosporin A; CYC, cyclophosphamide; AZA, azathioprine; cortico's, corticosteroids.

31 32 33 34 36

Bensen67 Kavanaugh68 Moreland69 Kerstens70 Wiesenhutter71

Therapy-resistance management 741

742 E.-J. J. A. Kroot et al

de®nition. We recommend that these provisional criteria should minimally include the following characteristics before patients can be considered as having refractory RA: . 53 DMARDs prescribed in treatment . the DMARDs used should minimally include methotrexate 15 mg/weekly and sulphasalazine 2000 mg/day for a period of 6 months, stopped because of lack of ecacy (Disease Activity Score (DAS) improvement 40.6 in comparison with pretreatment DAS4) or stopped for less than 6 months because of toxicity . persistent disease activity as measured by DAS 43.7 4 Practical current management of refractory RA As no exact de®nition of the term `refractory' RA is yet available, for practical purposes, patients will be considered to have refractory RA when they have been unsuccessfully treated with at least 2 DMARDs, preferably including methotrexate and sulphasalazine.3 For these patients several management strategies can be considered, including dose-increase of DMARD, combination therapy, adding corticosteroids, and in case of mono-articular involvement local treatment with corticosteroids or radiochemicals. Increasing the dose above the standard dosage regimen is in most patients the ®rst option. This possibility has been extensively carried out with methotrexate5 and sulphasalazine.6 In recent years weekly doses of 25±30 mg have been successfully used without increases of toxicity, whereas before 1990 the upper limit of weekly methotrexate was 15 mg. In the future the maximum weekly dosage of methotrexate prescribed may be increased to over 30 mg.7 In sulphasalazine treated patients, receiving doses up to 2000 mg, increases to 3000 mg daily have led to improvement in approximately 50% of the patients.6 These dose increases may be followed by combination therapy. This therapeutic option was not very successful in the past, but recently has been the subject of a number of well-designed studies.8,9,84 In patients with recent onset RA, step-down bridge therapy, including corticosteroids, leads to enhanced ecacy, at acceptable or low toxicity.10 However, clinical improvement does not have to correlate with lower radiological joint damage. In patients with severe RA, cyclosporin improves a suboptimal clinical response to methotrexate9 and the triple combination of methotrexate, sulphasalazine, and hydroxychloroquine appears to be clinically more e€ective than the individual components.8 An important and interesting ®nding in these studies was that no signi®cant increased toxicity pro®le was observed in combination therapy. In a parallel-designed study of early RA patients, comparing the combination of methotrexate and sulphasalazine with each of these drugs alone, showed the combination was no more e€ective, although some variables were found in favour of the combination.11 For practical purposes the step-up strategy is recommended. However, a recent study suggested that the step-down or parallel strategy may be more useful than step-up strategy, particularly in clinical trials.10 The symptomatic e€ects in refractory RA patients are probably dependent on corticosteroid use, including both low daily dose or depot (bridging) therapies, which are often included in the treatment of these patients.84 Management of resistant synovitis Persistent mono- or oligo-articular synovitis; despite treatment with anti-in¯ammatory drugs, disease modifying agents and several intra-articular corticosteroid injections

Therapy-resistance management 743

occurs only in a limited number of patients with rheumatoid arthritis. Treatment options for these patients are often limited to radiochemical or surgical synovectomies, or even total joint replacement(s), as persistent synovitis often leads to severe articular damage.12 In most cases surgical synovectomy is still considered when patients have not responded well to radiochemical synovectomy. Radiochemical synovectomy is occasionally used in conjunction with arthroscopic synovectomy.12 The exact mechanism of action of radiochemical synovectomies is unknown. The most likely explanation for the anti-in¯ammatory e€ect of radiation on the synovitis, is the resorption of the radiopharmacon by the macrophages of the synovial membrane. The choice of a speci®c radiopharmacon depends on the size of the joint.13 Better results are obtained when no radiological damage is observed before treatment.13,14 These radiochemical synovectomies have now been used for more than 50 years. However, few placebo controlled clinical trials have been designed, in which the ecacy, safety, choice of agent and ways of administration of these radiochemicals have been studied.12 In a recent review the most important studies concerning radiochemical synovectomies in the last 10 years have been reported.12 In these studies a great variety of radiopharmaceutical preparations have been reported, including yttrium-90 (Y-90), erbium-169 (Er-169), rhenium-186 (Re-186), gold-198 and dysprosium-165.15 Y-90 with citrate or silicate colloid for the larger joints and Er-169 for the smaller joints, both pure beta-ray emitters, are the most utilized agents at present.16 Y-90 is mostly used in larger joints, such as the knee, because of its high tissue penetrance.13 Er-169, an agent with short tissue penetrance, is commonly used in smaller joints such as the ®ngers. Both radio-isotopes have short half-lives, desirable for reducing radiation damage. Trials examining the ecacy of radiochemical synovectomies followed up for at least 6 months reported di€erent response rates, ranging from 32±95%.12 No serious side e€ects have been reported in 50 years of treatment. Post-injection extra-articular leakage to lymph nodes, liver, lung and spleen have been reported, but was not substantial. In most cases only 1% of the injected dose leaks to other organs. Chromosomal abnormalities and changes in peripheral lymphocytes have been reported, but no increases in cancer were observed during the 50 years of observation.12 So radiochemical synovectomy is a generally safe treatment option for persistent synovitis refractory compared with other conventional antirheumatic treatments, and may be considered as a less costly alternative to surgery and having a similar long-term outcome. However, only temporary symptomatic relief is obtained, whereas disease progression is una€ected.17 Surgical synovectomy is indicated when the synovitis has not responded to conventional therapy, including intra-articular corticosteroid injections and radiochemical synovectomy. Surgical synovectomies are in general considered contra-indicated when two or more joints are involved.18 Most studies concerning surgical synovectomies, reported synovectomies for the wrist, elbow and knee joints.18 However, these studies were focused on local ®ndings, like decrease of swelling, decrease of pain and range of motion, but did not report whether RA activity in other joints was controlled as well. MANAGEMENT OF EXTRA-ARTICULAR COMPLICATIONS Management of SjoÈgren's syndrome SjoÈgren's syndrome is a chronic in¯ammatory disease characterized by the in®ltration of exocrine glands by lymphocytes and plasma cells. Traditionally, SjoÈgren's syndrome

744 E.-J. J. A. Kroot et al

is divided into primary (no association with another connective disease present) and secondary forms (association with connective tissue disease such as RA, systemic lupus erythematosus, systemic sclerosis, vasculitis and primary biliary cirrhosis). Sicca symptoms develop as a result of glandular destruction, and are characterised by ocular, oral and vaginal mucosal dryness. Recently developed diagnostic criteria for SjoÈgren's syndrome may further facilitate its diagnosis.19 The prevalence of SjoÈgren's syndrome in RA patients is about 5%.19,20 Dilation of the bulbar conjunctival vessels and pericorneal injection is often observed by physical examination. Rose-bengal staining for detecting damage to the epithelium of the cornea by slit lamp examination may be used to con®rm the diagnosis of keratoconjunctivitis (KCS). For testing tear secretion by the lacrimal glands, Schirmer's tear test is used. Other options for detecting salivary gland involvement include sialometry, sialography and salivary scintigraphy. However, the only reliable method for diagnosing SjoÈgren's syndrome for certain, is minor salivary gland biopsy. As SjoÈgren's syndrome is a disease with a wide clinical spectrum, treatment consists mainly of the application of topical preparations to replace ¯uids. Arti®cial tear drops may be used as often as necessary. For protecting the cornea soft contact lenses may be helpful. Management of xerostomia is less satisfactory and largely palliative. Dry food, drugs with anticholinergic side e€ects, smoking and use of sugar (essentially in the prevention of dental disease) should be avoided. Lubricant jellies are useful in the treatment of vaginal dryness and dry skin with moisturizing lotions. The in¯uence of drug treatment on the disease course of SjoÈgren's syndrome remains unsatisfactory, but systemic corticosteroids and alkylating agents are recommended for progressive extraglandular disease. One should always be aware of the development of lymphoproliferative disorders, particularly in the salivary glands and the lungs, although this is a rare complication in SjoÈgren's syndrome patients and occurs mainly in its primary phase.21 Management of lung involvement In RA the lungs may be involved in several ways, developing pleuritis, pleural e€usion, parenchymal nodules, interstitial involvement, rheumatoid pneumoconiosis, pneumonitis and airway obstruction, and, more rarely, vasculitis.22 Pleural involvement, usually asymptomatic, is the most common pulmonary manifestation of RA and occurs in many cases concurrently with pulmonary nodulosis or interstitial disease.22 The course and clinical features of pulmonary ®brosis in RA patients are similar to those of idiopathic pulmonary ®brosis. Approximately 20±40% of RA patients have symptoms due to pleural involvement or restrictive abnormalities, but the diagnosis of rheumatoid lung-disease is made in only 1±5% of RA patients.22 However, in post mortem studies prevalence rates of 40±75% have been reported. In addition to these primary pulmonary manifestations of RA, environmental exposures, infectious agents and toxic consequences of drug treatment are also associated with the development of pulmonary disease in RA patients.23 Lower obstructive pulmonary disease in RA patients includes bronchi-ectasis and obliterative bronchiolitis, which may be idiopathic or induced by d-penicillamine or intramuscular gold compounds.22 Upper airway obstruction is less frequent although awareness of crico-artenoid involvement or vocal cord ®xation should always be kept in mind. Methotrexate- (MTX) induced pneumonitis is an uncommon, but severe complication of methotrexate treatment, with aspeci®c clinical manifestations. No consensus exists on whether pre-existing

Therapy-resistance management 745

pulmonary disease predisposes these patients to methotrexate pneumonitis.22 The most common clinical ®ndings include acute onset of dyspnoea, fever, non-productive cough and cyanosis. The di€erential diagnosis in MTX-treated patients should therefore include MTX-pneumonitis, rheumatoid lung disease, and pulmonary infection or emboli.24 As the exact mechanism of MTX-induced lung pathology remains unclear, the approach to patients with suspected MTX-related lung pathology consists of MTX discontinuation, supportive therapy, comprehensive diagnostic procedures to exclude infection (including bronchoalveolar lavage (BAL) analysis), empirical anti-microbial treatment and eventually intravenous corticosteroids until clinical and radiological improvement appear.24 So physicians must be careful not to attribute all lung diseases to the underlying disease (RA). Evaluation of pulmonary involvement in RA patients is usually done by chest radiographs, lung function tests and BAL.25 Results from a recent study, in which BAL was performed in RA patients, showed that an altered balance of immunocompetent cells is not only seen in the joints, but also in the lungs. This altered balance appeared to be more distinct if associated with clinical signs and symptoms.26 Unfortunately, controlled studies concerning pulmonary diseases in RA patients do not exist at present. It appears that in general, treatment of RA is often all that is required for controlling lung involvement. Further treatment of pulmonary diseases remains empirical and relies on physician judgement based on the clinical pro®le of each patient. At the moment there are no generally accepted markers of increased susceptibility to lung disease in RA patients, which complicates the assessment of whether, for example, bronchoscopy is indicated to assess lung involvement in RA patients. In some patients high doses of corticosteroids are bene®cial, but in most patients an immunosuppressive agent should be introduced to reduce progression. So, when aggressive therapy is necessary, pulses of methylprednisolone are recommended to favour the lowest dose of oral prednisone. Thereafter, azathioprine, an immunosuppressor, should be considered. Combination of prednisone and azathioprine has been reported to have a better ecacy and a lower toxicity pro®le than the combination of prednisone and cyclophosphamide.22 The latter drug should be considered when azathioprine is ine€ective or causes severe side e€ects.22 Management of rheumatoid vasculitis Vasculitis, characterised by destruction of the blood vessels through in¯ammatory processes, is often ®rst noticed by tissue ischaemia.27 This rare extra-articular manifestation of rheumatoid arthritis may be devastating because of gangrene and internal organ damage.28 Rheumatoid vasculitis has a low frequency ranging from 1±5%, but in post mortem examinations the frequency has been reported to be much higher, ranging from 1±25%. The vascular lesions observed in the vessels result from immunemediated damage of the vascular endothelium. As RA synovitis seems to be driven by macrophage activation, induced by Th1-cells, vasculitis (but also nodule formation and Felty's syndrome) is characterized by B-cell overactivity, immune complex formation and complement consumption, suggesting that Th2-cells are involved in the pathogenesis of vasculitis and other extra-articular manifestations of RA.29 Several studies have shown that rheumatoid vasculitis must be subdivided into three clinicopathological groups according to the size of vessel30: 1. obliterative endarteritis with intimal proliferation and thrombosis of digital vessels without signs of in®ltrate

746 E.-J. J. A. Kroot et al

2. in¯ammation of small venules or small arterioles in the skin, with leucocytoclasis as the most characteristic morphological feature 3. necrotizing arteritis involving small and medium sized arteries which can a€ect visceral organs or peripheral nerves and is clinically indistinguishable from polyarteritis nodosa, with ®brinoid necrosis as the main clinical feature Rheumatoid vasculitis covers a wide clinical spectrum ranging from benign nailfold lesions and rash and skin eruptions to infarctions of visceral organs, gangrene and/or mononeuritis multiplex. However, the skin and nerves are usually the only a€ected organs.28 The typical onset of rheumatoid vasculitis consists of a mild symmetric peripheral sensory neuropathy a€ecting the lower legs, in some patients resulting in ischaemic ulcers in addition to petechiae or purpura. Biopsies of the vascular lesions are needed for diagnostic certainty. Patients with rheumatoid vasculitis are more often men, have more frequently increased rheumatoid factor concentrations, joint erosions and subcutaneous nodules compared with RA patients without vasculitis.31 The choice of treatment depends on both the presence and severity of the speci®c clinical symptoms and on the type of vessel involvement.30 A great variety of drugs has been used in the treatment of severe rheumatoid vasculitis patients with systemic manifestations. Several treatment modalities, including corticosteroids, azathioprine, cyclophosphamide and less frequently methotrexate and plasma exchange, have been used in several controlled studies.30 However, in most studies the number of treated patients was limited. In patients with only mild rheumatoid arthritis, skin rash, super®cial ulcers and nailfold lesions, no aggressive immunosuppressive cytotoxic therapy is needed, as they have a better prognosis compared to patients with severe or systemic rheumatoid vasculitis. In patients so a€ected, immediate aggressive immunosuppressive therapy is recommended to improve survival and clinical features. Corticosteroids appeared to be of limited bene®t in severe rheumatoid vasculitis. Combination therapies of corticosteroids and cytostatic drugs have been reported to be more e€ective in several trials.85 Thus, combination therapy of intravenous cyclophosphamide and methyl-prednisone has been so far the treatment of ®rst choice. This therapy has shown clear clinical bene®t, but not a better survival. Combination therapy of azathioprine and prednisone induced remission in most of the patients with severe rheumatoid vasculitis, whereas few adverse events and a low rate of relapse was noticed.85 In addition, due to the promising results of this combination therapy, patients with less severe rheumatoid vasculitis (vasculitis which was only restricted to the skin) were treated with azathioprine and prednisone or conventional DMARDs. Results of this study showed that there is no need for these patients to be treated immediately with immunosuppressive cytotoxic drugs. Long-term controlled trials have to be developed for patients with severe rheumatoid vasculitis in order to determine which combination of high-dose (intermittent pulses of) oral cyclophosphamide or azathioprine in combination with high dose corticosteroids is the most e€ective and safe treatment of ®rst choice.85 Management of nodulosis Rheumatoid nodulosis appears in 20±25% of RA patients, more commonly between age 30±50 years and more frequently in men.32 Most patients have a positive test for rheumatoid factor, develop a more severe disease, report more other extra-articular manifestations and have lower remission rates.32 Nodules are often detected on pressure areas, including elbows, sacrum, ®nger joints, occipital scalp and Achilles

Therapy-resistance management 747

tendon. These nodules are ®rm and often adherent to the periosteum. A focal central area with ®brinoid necrosis, surrounded by ®broblasts, is found by histological examinations, resulting from small vessel vasculitis.33 Nodules are not only a diagnostic feature of RA, but may also serve as a site for local infection or as a portal for systemic infection, causing septic arthritis. Vasculitis of the overlying skin may also contribute to this severe complication. RA patients developing septic manifestations have to be thoroughly examined for nodules. Clinically, rheumatoid nodules have to be di€erentiated from similar nodules occurring in rheumatic fever, tophaceous gout, granuloma annulare and amyloidosis.32 Therapeutic strategies directed against rheumatoid nodulosis remain controversial. Subcutaneous nodules may regress with conventional DMARD treatment, usually with RA improvements also. Nevertheless, some studies reported that treatment with gold salts, D-penicillamine or hydroxychloroquine have only produced articular improvements, without decrease in size or appearance of new rheumatoid nodules.32 Methotrexate has been reported to exacerbate nodules in some patients, despite e€ective suppression of the disease activity. Additional treatment with hydroxychloroquine (400 mg/day) has led to a signi®cant reduction in number and size of these methotrexate induced, accelerated rheumatoid nodules between 3 and 10 months after starting this combination therapy in several case reports.33 Another exciting new option in the treatment of methotrexate induced nodulosis is inhibition of adenosine A1-receptors, while still potentiating the A2-mediated anti-in¯ammatory e€ects of methotrexate on synovitis.34 Management of cervical subluxation The cervical spine is commonly involved in patients with severe RA, particularly at the C1-C2 level, the atlanto±axial joint, resulting in anterior, horizontal subluxation of the atlas or in atlanto±axial impaction (cranial settling). Subluxations at lower levels are less frequently found.35 In almost 30% of the patients with severe RA cervical subluxation may be found.36 More patients are male, have high titres of rheumatoid factor, and have more radiological damage in peripheral joints already at an early stage of the disease. Radiographs of the cervical spine are strongly recommended in these patients.37 The in¯ammatory process causes ligamentous distension and rupture, as well as erosive changes, responsible for instability of the cervical spine. The consequences of cervical subluxation are non-traumatic dislocations and compression of the spinal cord. When cervical cord compression occurs due to atlanto±axial subluxation, patients have an estimated risk of 10% of sudden death due to medullary compression. Alarm signs include intractable neck pain radiating to the occiput, sensory and pyramidal tract weakness of the arms and legs, a `marble' sensation in limbs and trunk, jumping legs and a disturbed bladder function.38 Neurosurgical intervention in these patients is often indicated.35 Lateral radiographs of the upper cervical spine in maximal ¯exion are commonly used as a diagnostic tool for C1-C2 subluxation. Magnetic resonance imaging (MRI)-myelography is another diagnostic option in assessment of spinal cord damage in the craniocervical region.35 Most of the patients have only symptoms of occipital headache. For these patients a neck collar, NSAIDs if necessary and neck traction may be sucient. However, a surgical approach in patients with intractable neck pain and instability without neurological signs is recommended by others.38 Indications for surgical stabilisation are cervical cord compression con®rmed by MRI and neurological signs. Surgical stabilisation includes fusion of C1-C2 spinous processes with the occiput and

748 E.-J. J. A. Kroot et al

some lower vertebrae by transarticular facet screw ®xation. In some cases fusion with cannulated screws is sucient.36 Patients with these severe symptoms, who refuse or cannot be operated on (because of too great an operation risk), have a poor 1-year prognosis. But when an operation is undertaken, improved long-term outcomes have been reported for both neurological symptoms and symptomatic pain relief. However, poor bone quality, impaired wound healing, susceptibility to infections and joint disability often complicate the peri- and post-surgical long-term improvements.39 Recurrent subluxation, mostly at the subaxial levels, is observed after some years, but the life expectancy is approximately 5 years in 50% of the patients. So at present early surgery is recommended, as correction of symptomatic atlanto±axial subluxation may prevent progression of instability.36 FUTURE MANAGEMENT OF REFRACTORY RHEUMATOID ARTHRITIS At present, one of the most exciting areas in treatment of RA patients, is the development of target-oriented agents, including monoclonal antibodies. These agents are now being tested in a number of clinical trials where they are directed at individual cells or at mediators, in particular cytokines, generated by these cells. Because of the promising results of these trials, performed in RA patients with more severe disease, some of these biological agents will soon ®nd their way to clinical practice. As biological agents against tumour necrosis factor alpha (TNF-a) have been proven to be very relevant in the disease process in RA, treatment with TNF-a blocking agents seems so far the most promising and successful agent in producing clinical improvement. Up to now, these biologicals have only been used in a small number of patients followed up for not more than 3±4 years. So it remains unclear whether the long-term outcome, in particular in terms of structural joint damage and side e€ects, is also in¯uenced by these interventions. Therapy with the CD4-monoclonal antibody has not been very successful up to now, but this therapeutic strategy certainly will be studied further.40 As there is still no clear evidence whether the in¯ammatory and destructive pathways are dissociated or not, this piece of information concerning the pathogenesis of RA still has to be elucidated. If it turns out that both pathways are (partly) uncoupled, the goals of therapy should be both symptom (in¯ammation) and Practice points . Criteria for refractory RA should minimally include: 1. 53 DMARDs prescribed in treatment; 2. these DMARDs should minimally include methotrexate 15 mg/weekly and sulphasalazine 2000 mg/day for a period of 6 months, stopped because of lack of ecacy (DAS improvement 40.6 in comparison with pre-treatment DAS4) or less than 6 months stopped because of toxicity; 3. persistent disease activity as measured by DAS 4 3.7 4 . only a small number of RA patients will become treatment resistant during their disease course . physicians should be more aware of the occurrence of extra-articular complications in RA patients, as their prevalence is underestimated

Therapy-resistance management 749

Research agenda . further characterisation of the term refractory RA to improve trial comparability . the promising development of target-oriented agents in the treatment of RA patients has created an urgent need for well-de®ned guidelines indicating how and when to use these agents . placebo-controlled clinical trials are needed, in which ecacy and safety of radiochemical and surgical synovectomies are studied, to improve treatment of resistant mono- or oligo-articular synovitis disease (destruction) modi®cation. Therefore, well-de®ned guidelines are necessary, how and when these biological agents should be used; whether as induction therapy, as part of combination therapy, as chronic maintenance monotherapy or as a kind of bridging therapy. At present only one study has presented results on combination therapy of conventional DMARDs and biological agents, but further well-de®ned studies concerning this combination therapy are still awaited.41 So in the future a careful de®ned clinical programme has to be developed in which the goals of these target-oriented treatments are exactly de®ned. Other possible useful treatment options in the future include gene therapy42 stem-cell treatment43 and oral tolerance induction.44 REFERENCES * 1. van Riel PL. Provisional guidelines for measuring disease activity in clinical trials on rheumatoid arthritis [editorial]. British Journal of Rheumatology 1992; 31: 793±794. * 2. Felson DT, Anderson JJ, Boers M et al. The American College of Rheumatology preliminary core set of disease activity measures for rheumatoid arthritis clinical trials. The Committee on Outcome Measures in Rheumatoid Arthritis Clinical Trials. Arthritis and Rheumatism 1993; 36: 729±740. * 3. Felson DT, Anderson JJ & Meenan RF. Use of short-term ecacy/toxicity trade o€s to select second-line drugs in rheumatoid arthritis. A meta-analysis of published clinical trials. Arthritis and Rheumatism 1992; 35: 1117±1125. * 4. van Riel PL, van Gestel AM & van de Putte LB. Development and validation of response criteria in rheumatoid arthritis: steps towards an international consensus on prognostic markers. British Journal of Rheumatology 1996; 35 (supplement 2): 4±7. 5. van Ede AE, Laan RF, Blom HJ et al. Methotrexate in rheumatoid arthritis: an update with focus on mechanisms involved in toxicity. Seminars in Arthritis and Rheumatism 1998; 27: 277±292. 6. van Riel PL, van Gestel AM & van de Putte LB. Long-term usage and side-e€ect pro®le of sulphasalazine in rheumatoid arthritis. British Journal of Rheumatology 1995; 34 (supplement 2): 40±42. 7. Shiroky JB, Neville C & Skelton JD. High dose intravenous methotrexate for refractory rheumatoid arthritis. Journal of Rheumatology 1992; 19: 247±251. 8. O'Dell JR, Haire CE, Erikson N et al. Treatment of rheumatoid arthritis with methotrexate alone, sulfasalazine and hydroxychloroquine, or a combination of all three medications [see comments]. New England Journal of Medicine 1996; 334: 1287±1291. 9. Tugwell P, Pincus T, Yocum D et al. Combination therapy with cyclosporine and methotrexate in severe rheumatoid arthritis. The Methotrexate±Cyclosporine Combination Study Group [see comments]. New England Journal of Medicine 1995; 333: 137±141. 10. Verhoeven AC, Boers M & Tugwell P. Combination therapy in rheumatoid arthritis: updated systematic review. British Journal of Rheumatology 1998; 37: 612±619. 11. Haagsma CJ, van Riel PL, de Jong AJ & Van de Putte LB. Combination of sulphasalazine and methotrexate versus the single components in early rheumatoid arthritis: a randomized, controlled, double-blind, 52 week clinical trial. British Journal of Rheumatology 1997; 36: 1082±1088.

750 E.-J. J. A. Kroot et al 12. Sholter D & Davis P. Radiochemical synovectomy [editorial]. Scandinavian Journal of Rheumatology 1997; 26: 337±341. 13. Stucki G, Bozzone P, Treuer E et al. Ecacy and safety of radiation synovectomy with yttrium-90: a retrospective long-term analysis of 164 applications in 82 patients [see comments]. British Journal of Rheumatology 1993; 32: 383±386. 14. Asavatanabodee P, Sholter D & Davis P. Yttrium-90 radiochemical synovectomy in chronic knee synovitis: a one year retrospective review of 133 treatment interventions. Journal of Rheumatology 1997; 24: 639±642. 15. Glunie G & Ell PJ. A survey of radiation synovectomy in Europe, 1991±1993. European Journal of Nuclear Medicine 1995; 22: 970±976. 16. Aguilera S, Pizzi T & Donoso I. Sinovectomia radioactiva con yttrium 90 y rhenium 186, en artritis reumatoidea, seguimiento a largo plazo y efecto en la membrana sinovial. [Radiation synovectomy with yttrium 90 and rhenium 186 in rheumatoid arthritis, long term follow-up and e€ects on synovial membrane.] Revista Medica de Chile 1994; 122: 1283±1288. 17. Cruz EC & Wilke WS. Innovative treatment approaches for rheumatoid arthritis. Non-surgical synovectomy. BaillieÁres Clinical Rheumatology 1995; 9: 787±801. 18. Ayral X, Bonvarlet JP, Simonnet J et al. Arthroscopy-assisted synovectomy in the treatment of chronic synovitis of the knee. Revue du Rhumatisme 1997; 64: 215±226 (English edition). 19. Vitali C, Bombardieri S, Moutsopoulos HM et al. Assessment of the European classi®cation criteria for SjoÈgren's syndrome in a series of clinically de®ned cases: results of a prospective multicentre study. The European Study Group on Diagnostic Criteria for SjoÈgren's Syndrome. Annals of Rheumatic Diseases 1996; 55: 116±121. 20. Moutsopoulos HM & Kordossis T. SjoÈgren's syndrome revisited: autoimmune epithelitis [editorial]. British Journal of Rheumatology 1996; 35: 204±206. 21. Matsuo T, Kono R, Matsuo N et al. Incidence of ocular complications in rheumatoid arthritis and the relation of keratoconjunctivitis sicca with its systemic activity. Scandinavian Journal of Rheumatology 1997; 26: 113±116. *22. Anaya JM, Diethelm L, Ortiz LA et al. Pulmonary involvement in rheumatoid arthritis. Seminars in Arthritis and Rheumatism 1995; 24: 242±254. 23. Byrd SL, Case BA & Boulware DW. Pulmonary manifestations of rheumatic disease. Postgraduate Medicine 1993; 93: 149±151, 154±156, 159±161. 24. Barrera P, Laan RF, van Riel PL et al. Methotrexate-related pulmonary complications in rheumatoid arthritis. Annals of Rheumatic Diseases 1994; 53: 434±439. 25. Scherak O, Kolarz G, Popp W et al. Lung involvement in rheumatoid factor-negative arthritis. Scandinavian Journal of Rheumatology 1993; 22: 225±228. 26. Kolarz G, Scherak O, Popp W et al. Bronchoalveolar lavage in rheumatoid arthritis. British Journal of Rheumatology 1993; 32: 556±561. 27. Voskuy AE, Zwinderman AJ, Westedt ML et al. The mortality of rheumatoid vasculitis compared with rheumatoid arthritis [see comments]. Arthritis and Rheumatism 1996; 39: 266±271. 28. Halverson PB. Extra-articular manifestations of rheumatoid arthritis. Orthopaedic Nursing 1995; 14: 47±50. 29. Snowden N & Kay RA. Immunology of systemic rheumatoid disease. British Medical Bulletin 1995; 51: 437±448. 30. Lie JT. Diagnostic histopathology of major systemic and pulmonary vasculitic syndromes. Rheumatic Disease Clinics of North America 1990; 16: 269±292. 31. Voskuyl AE, Zwinderman AH, Westedt ML et al. Factors associated with the development of vasculitis in rheumatoid arthritis: results of a case-control study. Annals of Rheumatic Diseases 1996; 55: 190±192. 32. Goni MA, Scheines EJ, Paira SO et al. Rheumatoid nodulosis: a puzzling variant of rheumatoid arthritis. Clinical Rheumatology 1992; 11: 396±401. 33. Combe B, Didry C, Gutierrez M et al. Accelerated nodulosis and systemic manifestations during methotrexate therapy for rheumatoid arthritis. European Journal of Medicine 1993; 2: 153±156. 34. Merrill JT, Shen C, Schreibman D et al. Adenosine A1 receptor promotion of multinucleated giant cell formation by human monocytes: a mechanism for methotrexate-induced nodulosis in rheumatoid arthritis. Arthritis and Rheumatism 1997; 40: 1308±1315. 35. Romanowski CA, Nisar M & Nakielny RA. Atlanto-occipital subluxation in rheumatoid arthritis demonstrated by magnetic resonance imaging. British Journal of Rheumatology 1995; 34: 787±789. 36. McRorie ER, McLoughlin P, Russell T et al. Cervical spine surgery in patients with rheumatoid arthritis: an appraisal. Annals of Rheumatic Diseases 1996; 55: 99±104. 37. Paimela L, Laasonen L, Kankaanpaa E & Leirisalo RM. Progression of cervical spine changes in patients with early rheumatoid arthritis. Journal of Rheumatology 1997; 24: 1280±1284.

Therapy-resistance management 751 38. Rawlins BA, Girardi FP & Boachie AO. Rheumatoid arthritis of the cervical spine. Rheumatic Disease Clinics in North America 1998; 24: 55±65. 39. Gar®n SR & Ahlgren BD. Current management of cervical spine instability. Current Opinion in Rheumatology 1995; 7: 114±119. 40. van der Lubbe PA, Dijkmans BA, Markusse HM et al. A randomized, double-blind, placebo-controlled study of CD4 monoclonal antibody therapy in early rheumatoid arthritis. Arthritis and Rheumatism 1995; 38: 1097±1106. *41. Maini RN, Breedveld FC, Kalden JR et al. Therapeutic ecacy of multiple intravenous infusions of antitumor necrosis factor alpha monoclonal antibody combined with low-dose weekly methotrexate in rheumatoid arthritis [see comments]. Arthritis and Rheumatism 1998; 41: 1552±1563. 42. Bakker AC, Joosten LA, Arntz OJ et al. Prevention of murine collagen-induced arthritis in the knee and ipsilateral paw by local expression of human interleukin-1 receptor antagonist protein in the knee. Arthritis and Rheumatism 1997; 40: 893±900. 43. Tyndall A & Gratwohl A. Haemopoietic stem and progenitor cells in the treatment of severe autoimmune diseases. Annals of Rheumatic Diseases 1996; 55: 149±151. 44. Sieper J, Kary S, Sorensen H et al. Oral type II collagen treatment in early rheumatoid arthritis. A double-blind, placebo-controlled, randomized trial. Arthritis and Rheumatism 1996; 39: 41±51. 45. Keysser G, Keysser C & Keysser M. Treatment of refractory rheumatoid arthritis with low-dose cyclophosphamide. Long-term follow-up of 108 patients. Zeitschrift fuÈr Rheumatologie 1998; 57: 101±107. 46. Moreland LW, Haverty TP, Wacholtz MC et al. Nondepleting humanized anti-CD4 monoclonal antibody in patients with refractory rheumatoid arthritis. Journal of Rheumatology 1998; 25: 221±228. 47. Schirmer M, Mur E, Pfei€er KP et al. The safety pro®le of low-dose cladribine in refractory rheumatoid arthritis. A pilot trial. Scandinavian Journal of Rheumatology 1997; 26: 376±379. 48. Isaacs JD, Burrows N, Wing M et al. Humanized anti-CD4 monoclonal antibody therapy of autoimmune and in¯ammatory disease. Clinical and Experimental Immunology 1997; 110: 158±166. *49. Moreland LW, Baumgartner SW, Schi€ MH et al. Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein [see comments]. New England Journal of Medicine 1997; 337: 141±147. 50. Moreland LW. Initial experience combining methotrexate with biologic agents for treating rheumatoid arthritis. Journal of Rheumatology 1996; 44 (supplement): 78±83. 51. Kovarik JM, Kurki P, Mueller E et al. Diclofenac combined with cyclosporine in treatment refractory rheumatoid arthritis: longitudinal safety assessment and evidence of a pharmacokinetic/dynamic interaction. Journal of Rheumatology 1996; 23: 2033±2038. 52. Maksymowych WP, Avina ZA, Luong M & Russell AS. High dose intravenous immunoglobulin (IVIg) in severe refractory rheumatoid arthritis: no evidence for ecacy. Clinical and Experimental Rheumatology 1996; 14: 657±660. 53. Moreland LW, Margolies G, Heck LW Jr et al. Recombinant soluble tumor necrosis factor receptor (p80) fusion protein: toxicity and dose ®nding trial in refractory rheumatoid arthritis. Journal of Rheumatology 1996; 23: 1849±1855. 54. Schrohenloher RE, Koopman WJ, Woodworth TG & Moreland LW. Suppression of in vitro IgM rheumatoid factor production by diphtheria toxin interleukin-2 recombinant fusion protein (DAB 486IL-2) in patients with refractory rheumatoid arthritis. Journal of Rheumatology 1996; 23: 1845±1848. 55. Lacki JK, Leszczynski P & Mackiewicz SH. Intravenous cyclophosphamide combined with methylprednisolone in the treatment of severe refractory rheumatoid arthritis: the e€ect on lymphocytes. Journal of Investigational Allergology and Clinical Immunology 1996; 6: 232±236. 56. Isaacs JD, Wing MG, Greenwood JD et al. A therapeutic human IgG4 monoclonal antibody that depletes target cells in humans. Clinical and Experimental Immunology 1996; 106: 427±433. 57. Sala F, Carotti M & Cervini C. Combination therapy of cyclosporine A with methotrexate or hydroxychloroquine in refractory rheumatoid arthritis. Scandinavian Journal of Rheumatology 1996; 25: 16±23. 58. Kanik KS, Yarboro CH, Naparstek Y et al. Failure of low-dose intravenous immunoglobulin therapy to suppress disease activity in patients with treatment-refractory rheumatoid arthritis. Arthritis and Rheumatism 1996; 39: 1027±1029. 59. Blanco R, Martinez TVM, Gonzalez GMA et al. Acute febrile toxic reaction in patients with refractory rheumatoid arthritis who are receiving combined therapy with methotrexate and azathioprine. Arthritis and Rheumatism 1996; 39: 1016±1020. 60. Isaacs JD, Manna VK, Rapson N et al. CAMPATH-1H in rheumatoid arthritis ± an intravenous doseranging study. British Journal of Rheumatology 1996; 35: 231±240. 61. Malaise MG, De Keyser P, De Backer M et al. The use of Sandimmun (cyclosporin A) in severe refractory rheumatoid arthritis: the Belgian experience. Clinical Rheumatology 1995; 14 (Supplement 2): 2(s).

752 E.-J. J. A. Kroot et al 62. Weinblatt ME, Maddison PJ, Bulpitt KJ et al. CAMPATH-1H, a humanized monoclonal antibody, in refractory rheumatoid arthritis. An intravenous dose±escalation study. Arthritis and Rheumatism 1995; 38: 1589±1594. 63. Moreland LW, Pratt PW, Mayes MD et al. Double-blind, placebo-controlled multicenter trial using chimeric monoclonal anti-CD4 antibody, cM-T412, in rheumatoid arthritis patients receiving concomitant methotrexate. Arthritis and Rheumatism 1995; 38: 1581±1588. 64. Matteson EL, Yocum DE, St Clair EW et al. Treatment of active refractory rheumatoid arthritis with humanized monoclonal antibody CAMPATH-1H administered by daily subcutaneous injection. Arthritis and Rheumatism 1995; 38: 1187±1193. 65. Moreland LW, Sewell KL, Trentham DE et al. Interleukin-2 diphtheria fusion protein (DAB486IL-2) in refractory rheumatoid arthritis. A double-blind, placebo-controlled trial with open-label extension. Arthritis and Rheumatism 1995; 38: 1177±1186. 66. Maksymowych WP, Avina ZA, Luong MH & Russell AS. An open study of pentoxifylline in the treatment of severe refractory rheumatoid arthritis [see comments]. Journal of Rheumatology 1995; 22: 625±629. 67. Bensen W, Tugwell P, Roberts RM et al. Combination therapy of cyclosporine with methotrexate and gold in rheumatoid arthritis (2 pilot studies). Journal of Rheumatology 1994; 21: 2034±2038. 68. Kavanaugh AF, Davis LS, Nichols LA et al. Treatment of refractory rheumatoid arthritis with a monoclonal antibody to intercellular adhesion molecule 1. Arthritis and Rheumatism 1994; 37: 992±999. 69. Moreland LW, Pratt PW, Bucy RP et al. Treatment of refractory rheumatoid arthritis with a chimeric anti-CD4 monoclonal antibody. Long-term followup of CD4‡ T cell counts. Arthritis and Rheumatism 1994; 37: 834±838. 70. Kerstens PJ, Boerbooms AM, Brummelkamp EP & van de Putte LB. Long-term methotrexate in refractory rheumatoid arthritis; concurrent use of prednisone possibly improves drug-survival [letter]. Clinical Rheumatology 1994; 13: 317±318. 71. Wiesenhutter CW, Irish BL & Bertram JH. Treatment of patients with refractory rheumatoid arthritis with extracorporeal protein A immunoadsorption columns: a pilot trial. Journal of Rheumatology 1994; 21: 804±812. 72. Flipo RM, Cortet B, Duquesnoy B & Delcambre B. Traitement de la polyarthrite rhumatoide refractaire par la ciclosporine. [Treatment of refractory rheumatoid arthritis with cyclosporine.] Revue de Medicine Interne 1994; 15: 166±173. 73. Benenson EV & Timina OB. Prospidine versus methotrexate pulse in highly active rheumatoid arthritis: a controlled 6-month clinical trial. Clinical Rheumatology 1994; 13: 54±59. 74. Polisson RP, Dooley MA, Dawson DV & Pisetsky DS. Interleukin-2 receptor levels in the sera of rheumatoid arthritis patients treated with methotrexate. Arthritis and Rheumatism 1994; 37: 50±56. 75. Moreland LW, Pratt PW, Sanders ME & Koopman WJ. Experience with a chimeric monoclonal anti-CD4 antibody in the treatment of refractory rheumatoid arthritis. Clinical and Experimental Rheumatology 1993; 11 (supplement 8): S153±S159. 76. Moreland LW, Bucy RP, Tilden A et al. Use of a chimeric monoclonal anti-CD4 antibody in patients with refractory rheumatoid arthritis. Arthritis and Rheumatism 1993; 36: 307±318. 77. Weusten BL, Jacobs JW & Bijlsma JW. Corticosteroid pulse therapy in active rheumatoid arthritis. Seminars in Arthritis and Rheumatism 1993; 23: 183±192. 78. Woodworth TG. Early clinical studies of IL-2 fusion toxin in patients with severe rheumatoid arthritis and recent onset insulin-dependent diabetes mellitus. Clinical and Experimental Rheumatology 1993; 11 (supplement 8): S177±S180. 79. Racadot E, Wijdenes J & Wendling D. Immunological follow-up of 17 patients with rheumatoid arthritis treated in vivo with an anti-T CD4‡ monoclonal antibody (B-F5) [see comments]. Clinical and Experimental Rheumatology 1992; 10: 365±374. 80. Isaacs JD, Watts RA, Hazleman BL et al. Humanised monoclonal antibody therapy for rheumatoid arthritis. Lancet 1992; 340: 748±752. 81. Tumiati B, Casoli P, Veneziani M & Rinaldi G. High-dose immunoglobulin therapy as an immunomodulatory treatment of rheumatoid arthritis. Arthritis and Rheumatism 1992; 35: 1126±1133. 82. Altman RD, Perez GO & Sfakianakis GN. Interaction of cyclosporine A and nonsteroidal antiin¯ammatory drugs on renal function in patients with rheumatoid arthritis. American Journal of Medicine 1992; 93: 396±402. *83. Sewell KL, Parker KC, Woodworth TG et al. DAB486 IL-2 fusion toxin in refractory rheumatoid arthritis. Arthritis and Rheumatism 1993; 36: 1223±1233. *84. Boers M, Verhoeven AC, Markusse HM et al. Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with sulphasalazine alone in early rheumatoid arthritis [see comments]. Lancet 1997; 350: 309±318. 85. Heurkens AH, Westedt ML & Breedveld FC. Prednisone plus azathioprine treatment in patients with rheumatoid arthritis complicated by vasculitis. Archives of Internal Medicine 1991; 151: 2249±2254.