Statistics from Altmetric.com
Systemic lupus erythematosus (SLE) has variable presentation, course and prognosis. The wide acceptance and popularity of the first EULAR recommendations for its management, published in 2008,1 prompted the subsequent development of specific recommendations regarding monitoring, neuropsychiatric and renal disease, as well as for pregnancy and women’s health in lupus.2–5 Since these publications, new data have emerged on treatment strategies and validated goals of treatment, alternative regimens of glucocorticoids (GC), ‘multitargeted’ therapy with the use of calcineurin inhibitors (CNIs) in lupus nephritis (LN), and the approval of the first biological therapy for SLE. These advances called for an update of the EULAR recommendations for lupus, capitalising on the strengths of and experience from the previous projects.6
After approval by the EULAR Executive Committee, the convenor (DB) and methodologist (GB) invited a Task Force to work on this update; two fellows (AF, MK) undertook the systematic literature review (SLR). The EULAR standardised operating procedures7 and the Appraisal of Guidelines Research and Evaluation instrument (AGREE II)8 were followed. Applying a Delphi-based methodology, 14 research questions were selected for SLR (online supplementary table 1). PubMed was screened using strings of relevant terms. Since this was an update of the previous 2007 recommendations, the SLR considered all English-language publications from 01/2007 until 12/2017, with two exceptions: (1) treatment of skin disease, where an unrestricted date search was performed and (2) renal disease, where search was limited to the period 01/2012–12/2017 (since the EULAR recommendations for LN were published in 2012). Pertinent articles, identified by manual search within the reference list of the originally retrieved publications, were also included. All retrieved items were refined based on article type, abstract, full-text content and number of included patients. The final level of evidence and grading of recommendations considered also the body of evidence that had informed the previous sets of EULAR recommendations for the management of SLE, as the convenor, methodologist and several of the panellists had also participated in the latter. A detailed presentation of the SLR results is given in online supplementary tables 2 and 3. Evidence was categorised based on the design and validity of available studies and the strength of the statements was graded (see online supplementary table 4). After rounds of discussions, the committee reached a consensus of 33 final statements, grouped in four broad categories (Goals of Treatment, Treatment of SLE, Specific manifestations, Comorbidities—table 1). Each Task Force member rated their agreement with each statement.
Results and discussion
SLE represents a challenge for the treating physician in terms of diagnosis and treatment. Its protean manifestations, often multisystem but occasionally limited to a few or single organ, have led some physicians to focus exclusively on evidence of serological autoimmunity (antinuclear and more specific autoantibodies), for a disease where diagnosis is clinical after excluding competing diagnoses. Monitoring of SLE through validated disease activity and chronicity indices, including physician global assessment (PGA), is recommended. For patients with severe disease, multidisciplinary care in dedicated lupus centres is desirable.9 Immunosuppressive (IS) therapy (for induction and maintenance of remission) is indicated in organ-threatening lupus.
Goals of treatment
To improve long-term patient outcomes, management should aim at remission of disease symptoms and signs, prevention of damage accrual and minimisation of drug side-effects, as well as improvement of quality of life.10 11 Complete remission (absence of clinical activity with no use of GC and IS drugs) is infrequent.12–16 To this end, newly defined low disease activity states (based on a SLEDAI score ≤3 on antimalarials, or alternatively SLEDAI ≤4, PGA≤1 with GC ≤7.5 mg of prednisone and well tolerated IS agents) have shown comparable rates with remission, regarding halting of damage accrual (OR 0.5–0.7 for increase in damage index) and prevention of flares.14 17–20 Accordingly, treatment in SLE should aim at remission or, if this state cannot be achieved, at low disease activity in all organ systems. In LN, therapy should aim at least partial remission (defined as ≥50% reduction in proteinuria [UPr] to subnephrotic levels and serum creatinine [SCr] within 10% from baseline) by 6–12 months; complete renal remission (proteinuria <500 mg/24 hours and SCr within 10% from baseline), however, may require longer treatment duration, often more than 12 and until 24 months. In monitoring renal response, reduction of UPr (to less than 0.8 g/day) following treatment is more important than residual haematuria.21 Patients with more severe proteinuria and longer-standing disease are less likely to respond or show more delayed responses.22 23
Prevention of disease flares is an additional milestone of SLE treatment. Although a universally accepted definition is lacking, most experts agree that a flare is a measurable increase in disease activity usually leading to change of treatment.24 Flares are common in the disease course and contribute significantly to organ damage accrual and worse outcome.17 25 26 Consistently reported risk factors for a higher disease flare rate include younger age at disease onset, no use of antimalarials, persistent generalised disease activity and serological activity (anti-dsDNA, low complement).27–31 Assessment of adherence to drug treatment, close monitoring and optimisation of disease control in these patients may reduce the risk for a flare.
Treatment of SLE
Hydroxychloroquine (HCQ) is recommended for all patients with SLE. There is evidence for multiple beneficial effects of HCQ in SLE,32 yet poor adherence to treatment is not uncommon.33–35 Drug blood levels can be used to assess compliance,33 35 but data are currently insufficient to recommend routine monitoring of drug levels. Concerns for retinal toxicity with long-term HCQ therapy led to the use of more sensitive screening techniques, with a prevalence of retinal abnormalities exceeding 10% after 20 years of continuous use.36 37 Major risk factors for retinopathy include duration of treatment (OR 4.71 for every 5 years of use), dose (OR 3.34 for every 100 mg daily dose), chronic kidney disease (adjusted OR 8.56) and pre-existing retinal or macular disease.37 Based on existing evidence suggesting that the risk of toxicity is very low for doses below 5 mg/kg real body weight, the daily dose should not exceed this threshold. Of note, efficacy of HCQ in lupus has been established in studies with a prescribed dose of 6.5 mg/kg/day, thus it remains to be confirmed whether a lower dose will have comparable clinical effects. Patients in long-standing remission may have their dose lowered, although no studies have formally addressed this strategy. The choice of quinacrine, an alternative antimalarial, can be considered in patients with cutaneous manifestations and HCQ-induced retinal toxicity.
GC can provide rapid symptom relief, but the medium to long-term aim should be to minimise daily dose to ≤7.5 mg/day prednisone equivalent or to discontinue them, because long-term GC therapy can have various detrimental effects including irreversible organ damage.38–41 Risks are substantially increased at continuous GC doses above 7.5 mg/day, with some studies suggesting that also lower doses might be harmful.17 42–44 To this end, two approaches can be considered: (1) use of pulses of intravenous methylprednisolone (MP) of various doses (depending on severity and body weight), which take advantage of the rapid non-genomic effects of GC45 and may allow for a lower starting dose and faster tapering of PO GC,46 47 and (2) early initiation of IS agents, to facilitate tapering and eventual discontinuation of oral GC (see below). High-dose intravenous MP (usually 250–1000 mg/day for 3 days) is often used in acute, organ-threatening disease (eg, renal, neuropsychiatric) after excluding infections.48
Immunosuppressive (IS) drugs
Consequent initiation of IS drugs facilitates a more rapid GC tapering and may prevent disease flares.49 The choice of agent depends on prevailing disease manifestation(s), patient age and childbearing potential, safety concerns and cost. Methotrexate (MTX) and azathioprine (AZA) should be considered in patients with poor symptom control after a trial with GC and HCQ or when HCQ alone is unlikely to be sufficient, due to the large experience gained with their use and their relatively safe profile.50 Published evidence is generally stronger for MTX than AZA, yet the latter is compatible with pregnancy contemplation. Mycophenolate mofetil (MMF) is a potent immunosuppressant with efficacy in renal and non-renal lupus (although not in neuropsychiatric disease).51–53 In a recent randomised, open-label trial in extrarenal SLE, enteric-coated mycophenolate sodium (EC-MPS) was superior to AZA in achieving remission and reducing flares.54 However, its teratogenic potential (needs to be discontinued at least 6 weeks before conceiving), along with its higher cost compared with AZA or MTX, poses a limitation towards universal recommendation in women of reproductive age with non-renal manifestations. Cyclophosphamide (CYC) can be considered in organ-threatening disease (especially renal, cardiopulmonary or neuropsychiatric) and only as rescue therapy in refractory non-major organ manifestations; due to its gonadotoxic effects, it should be used with caution in women and men of fertile age.55–57 Concomitant use of GnRH analogues attenuates the depletion of ovarian reserve associated with CYC therapy and is recommended in premenopausal patients with SLE.4 58 59 Information about the possibility of ovarian cryopreservation should be offered ahead of treatment. Other risks of CYC therapy such as malignancy and infections should also be considered.60 61
There is evidence to support beneficial effects of B-cell targeting agents in SLE.62–66 Belimumab should be considered in extrarenal disease with inadequate control (ongoing disease activity or frequent flares) to first-line treatments (typically including combination of HCQ and prednisone with or without IS agents), and inability to taper GC daily dose to acceptable levels (ie, maximum 7.5 mg/day). Patients with persistent disease may benefit from belimumab; more likely to respond are patients with high disease activity (eg, SLEDAI >10), prednisone dose >7.5 mg/day and serological activity (low C3/C4, high anti-dsDNA titres), with cutaneous, musculoskeletal and serological manifestations responding the most.67–69
Due to the negative results of randomised controlled trials (RCTs), RTX is currently only used off-label, in patients with severe renal or extrarenal (mainly haematological and neuropsychiatric) disease refractory to other IS agents and/or belimumab, or in patients with contraindications to these drugs. As a general rule, more than one IS drug need to have failed prior to RTX administration,70–73 except perhaps for cases of severe autoimmune thrombocytopaenia and haemolytic anaemia, where RTX has demonstrated efficacy both in lupus and in patients with isolated immune thrombocytopaenia (ITP).74–76 In LN, RTX is typically considered following failure of first-line therapies (CYC, MMF) or in relapsing disease.70 77 More recently, a posthoc analysis of the LUNAR trial showed that complete B-cell depletion following RTX treatment in LN was associated with higher odds for complete response at 78 weeks.78
Figure 1 summarises the various drugs used in the treatment of SLE, according to disease severity stratification. Online supplementary table 5 outlines the recommended doses of the drugs mentioned in the manuscript.
A large body of evidence originates from studies in patients with cutaneous lupus erythematosus (CLE). Effective protection from ultraviolet exposure with broad-spectrum sunscreens and smoking cessation are strongly recommended.79–81 In atypical or refractory cases, a diagnostic skin biopsy should be considered. First-line treatment of skin disease includes topical agents (GC and/or CNIs) and antimalarials, with or without systemic GC (the latter at a starting dose depending on severity of skin involvement).82 83 HCQ is the antimalarial of choice over chloroquine due to its multiple beneficial effects and possibly lower risk for retinal toxicity;84 in cases of inadequate response or evidence of toxic retinopathy, quinacrine (mepacrine) may be used as an add-on or sequential therapy, respectively.85–87 Although quinacrine is currently unavailable in several countries worldwide, it is a useful alternative when available. There are no studies examining retinal toxicity of quinacrine with the newer, more sensitive screening techniques (visual fields or optical coherence tomography); however with current knowledge, retinopathy is not considered a side-effect of quinacrine.
A sizeable proportion (almost 40%) of patients will not respond to first-line treatment.86 88 In such cases, MTX can be added.50 89 Other agents include retinoids, dapsone and MMF or EC-mycophenolic acid.79 90 91 Belimumab and RTX have also shown efficacy in mucocutaneous manifestations of SLE, although these studies have not included a validated activity score for skin lesions; RTX may be less efficacious in chronic forms of skin lupus.62 92–94 Thalidomide is effective in various subtypes of cutaneous disease.95 96 Due to its strict contraindication in pregnancy, the risk for irreversible polyneuropathy, and the frequent relapses on drug discontinuation, it should be considered only as a ‘rescue’ therapy in patients who have failed multiple previous agents. A treatment algorithm for the various subtypes of CLE has been published by a European group of Dermatologists guided by the European Dermatology Forum in cooperation with the European Academy of Dermatology and Venereology.79
Neuropsychiatric disease (NPSLE)
Attribution of neuropsychiatric manifestations to SLE often requires a comprehensive, multidisciplinary approach to rule out mimics (infections, malignancy and others), taking into account the presence of risk (‘favouring ’) factors (type and timing of manifestation, presence of generalised, non-neurological disease activity, abnormal neuroimaging and cerebrospinal fluid analysis, positive antiphospholipid antibodies [aPL]),97 as well as confounding factors favouring alternative diagnoses.98 The use of validated attribution models may aid in the diagnostic process.99 100
Treatment of NPSLE depends on whether the underlying pathophysiological mechanism is presumed to be inflammatory or embolic/thrombotic/ischaemic.2 101 GC and/or IS agents should be considered in the former, while anticoagulant/antithrombotic treatment is favoured when aPL antibodies are present.102–107 Distinction between the two pathophysiological processes may not be easy in clinical practice, or the two processes may coexist in the same patient.2 Combination of IS and anticoagulant/antithrombotic therapy may be considered in these patients. Patients with SLE with cerebrovascular disease should be managed like the general population in the acute phase; in addition to controlling extra-CNS lupus activity, IS therapy may be considered in the absence of aPL antibodies and other atherosclerotic risk factors or in recurrent cerebrovascular events.108 In this context, neuroimaging and/or CSF studies may provide additional supporting evidence for IS therapy. Targeted symptomatic therapy is indicated according to the type of manifestation (eg, antipsychotics for psychosis, anxiolytics for anxiety disorder and so on).
Haematological manifestations frequently necessitating anti-inflammatory/IS treatment in patients with SLE include thrombocytopaenia and autoimmune haemolytic anaemia (AIHA). First-line treatment of significant lupus thrombocytopaenia (platelet count below 30 000/mm3) consists of moderate/high doses of GC in combination with IS agent (AZA, MMF or cyclosporine; the latter having the least potential for myelotoxicity) to facilitate GC-sparing. Initial therapy with pulses of intravenous MP (1–3 days) is encouraged. Intravenous immunoglobulin (IVIG) may be considered in the acute phase, in cases of inadequate response to high-dose GC or to avoid GC-related infectious complications.
Treatment of thrombocytopenia is typically lengthy and often characterised by relapses during GC tapering.109 In patients with no response to GC (ie, failure to reach a platelet count >50 000/mm3) or relapses, RTX should be considered, considering also its efficacy in ITP.74 76 110 CYC may also be considered in such cases. Thrombopoietin agonists or splenectomy should be reserved as last options.111 112 Autoimmune haemolytic anaemia (AIHA) is far less common than thrombocytopenia in SLE; its treatment follows the same principles regarding use of GC, IS drugs and RTX. Autoimmune leucopaenia is common in SLE but rarely needs treatment; careful work-up is recommended to exclude other causes of leucopaenia (especially drug-induced).
Patients at high risk of developing renal involvement (males, juvenile lupus onset, serologically active including positivity for anti-C1q antibodies)113–115 should be under vigilant monitoring (eg, at least every 3 months) to detect early signs of kidney disease. Following diagnosis, secured with a kidney biopsy, treatment of LN includes an initial induction phase, followed by a more prolonged maintenance phase. MMF and CYC are the IS agents of choice for induction treatment; low-dose CYC (Euro-Lupus regimen, online supplementary table 5) is preferred over high-dose CYC as it has comparable efficacy and lower risk of gonadotoxicity.57 116 117 Published data support the use of MMF and high-dose CYC (online supplementary table 5) in severe forms of LN associated with increased risk of progression into end-stage renal disease (reduced glomerular filtration rate, histological presence of fibrous crescents or fibrinoid necrosis, or tubular atrophy/interstitial fibrosis).118 119 An early significant drop in UPr (to ≤1 g/day at 6 months or ≤0.8 g/day at 12 months) is a predictor of favourable long-term renal outcome.21 117 120 MMF or AZA may be used as maintenance therapy, with the former associated with fewer relapses;121 122 the choice depends on the agent used for induction phase and on patient characteristics, including age, race and wish for pregnancy. In refractory or relapsing disease, RTX may be considered.
Following the EULAR recommendations for LN in 2012, several studies have been published regarding the use of CNIs to treat proliferative LN, either alone or in the form of a ‘multitarget therapy’ (combination of tacrolimus with MMF).123–127 These studies were performed almost exclusively in Asian populations and had short follow-up; hence, data have to be corroborated with longer duration studies in multiethnic populations. To this end, at present, CNIs may be considered as second-line agents for induction or maintenance therapy mainly in membranous LN, podocytopathy, or in proliferative disease with refractory nephrotic syndrome, despite standard-of-care within 3–6 months;128 129 in the latter case, they may be used alone or in combination with MMF, since small, observational studies have shown the CNI/MMF combination to be effective in disease refractory to standard therapy.130–132 Monitoring SCr and blood levels of CNI to avoid chronic drug toxicity is essential.
Antiphospholipid antibodies (aPL) and antiphospholipid syndrome (APS)
The presence of aPL is associated with thrombotic and obstetric complications and increased risk of damage accrual.133 134 In aPL carriers, a recent meta-analysis supported a protective role of low-dose aspirin for primary prophylaxis against thrombosis in the subgroup of aPL carriers who had SLE;135 however, in view of the potential bleeding hazard,136 137 it is not clear whether this should be applied to patients with lupus with any aPL antibodies or only to those carrying a high risk aPL profile (ie, triple aPL positivity, lupus anticoagulant or high titres of anticardiolipin antibodies).138 Patients with SLE with aPL may also receive additional anticoagulant treatment, such as low-molecular weight heparin, during high-risk periods for thrombosis (pregnancy or postoperatively), although no studies have formally addressed this question.
No studies have been performed exclusively on patients with SLE-APS, with several studies excluding secondary APS due to lupus. Thus, with current knowledge, treatment of APS in the context of SLE should not differ from treatment of primary APS. A recent randomised, open-label trial comparing rivaroxaban to warfarin in APS with triple aPL positivity (~21% of patients had SLE-APS) was prematurely terminated due to an excess of thromboembolic events in the rivaroxaban arm.139 Thus, in patients with SLE-APS, use of novel oral anticoagulants for secondary prevention should be avoided; however, they could potentially serve as an alternative option in selected patients (low-risk aPL profile, no history of arterial thrombotic events) with difficult to control international normalised ratio on warfarin, after balancing possible risks.
Risk of infection in SLE is associated with both disease-related and treatment-related factors; high-dose GC therapy, CYC, MMF and RTX are all associated with an increased risk for infection, while high disease activity, severe leucopaenia and presence of renal involvement (±hypogammaglobulinaemia in nephrotic syndrome) also contribute independently.48 140–143 Protection against infections should be proactive, focusing both on primary prevention, as well as timely recognition and treatment. Patients with lupus should receive vaccinations according to the EULAR recommendations for vaccination of patients with autoimmune rheumatic diseases.144 145 Immunisation against seasonal influenza and pneumococcal infection (both PCV13 and PPSV23) should be strongly considered, preferably during stable disease. Herpes zoster vaccination is now available for the general population, but a study in SLE has not been performed. Prompt diagnosis and treatment of sepsis is essential. To this end, validated scores such as the quick SOFA ([systolic blood pressure ≤100 mm Hg, respiratory rate ≥22/min, altered mental state with Glasgow coma scale <15]: the presence of ≥2 points near the onset of infection is associated with a greater risk of death or prolonged intensive care unit stay] may identify patients who are at greater risk for a poor outcome.146
SLE is an independent risk factor for cardiovascular disease (CVD), due to both traditional and disease-related risk factors, such as persistent disease activity, LN, presence of aPL and use of GC.147–149 Surrogate measures of atherosclerosis, such as carotid plaques, carotid intima media thickness (cIMT) and coronary artery calcium are frequently used to identify subclinical CVD in SLE.150 Low-dose aspirin may be considered for primary prevention of CVD, as it may reduce the risk for incident CVD in SLE (HR 0.24 in one retrospective study).151 152 However, this has to be viewed in light of recent large studies in diabetics and elderly showing that the benefits of aspirin for primary CVD prevention are counterbalanced by the larger bleeding hazard.136 153 The value of statins in SLE has been tested in RCTs, which failed to show a clear benefit over placebo, when cIMT was used a surrogate marker for CVD.154 155 Thus, routine use of statins is not recommended for all patients but should be considered on the basis of lipid levels and the presence of other traditional risk factors. Calculation of the 10-year CVD risk using the Systematic Coronary Risk Evaluation (SCORE, https://www.escardio.org/Education/Practice-Tools/CVD-prevention-toolbox/SCORE-Risk-Charts) is recommended,156 although the actual risk is underestimated in patients with SLE.
Certain points to consider and the research agenda suggested by the Task Force Members are reported in box 1. These points aim to improve the design of clinical studies in order to answer clinically important questions, for which the current ‘state-of-the-art’ is insufficient. In particular, data regarding the optimal duration and timing of discontinuation of therapy in both renal and extrarenal disease are scarce;157 for the former, recent studies support the value of a repeat kidney biopsy for the management of maintenance therapy, but more data are needed.158 159
Future research agenda in SLE
Targets of therapy
Exploration of a universally accepted level of residual disease activity, if remission cannot be achieved.
Existing therapies and disease monitoring
Efficacy of calcineurin inhibitor-containing treatment regimens in lupus nephritis (LN) in different racial/ethnic groups and at longer time points.
Usefulness of measurements of drug blood levels (hydroxychloroquine [HCQ], mycophenolate mofetil and so on).
Efficacy of quinacrine as immunomodulator in patients with HCQ-induced retinal toxicity.
Comparative trials of conventional immunosuppressive drugs with global and organ-specific result reporting.
Randomised trials testing lower cumulative dose glucocorticoid regimens versus conventional regimens.
Optimal treatment regimen of rituximab: regular versus on-demand.
Optimal duration of therapy and timing of discontinuation (renal and extrarenal disease).
Value of repeat kidney biopsy for monitoring LN and determination of clinical versus histological response to therapy.
Pathophysiology and Biomarkers
Susceptibility to develop systemic lupus erythematosus (SLE).
Involvement of particular organ systems over others, multisystem versus organ-dominant disease.
Response to specific therapeutic agents over others (pharmacogenetics, transcriptomics and so on).
Clinical trial design and new drug development
Optimisation of clinical trial design and study endpoints to maximise probability of new drug approval in SLE.
Handling of background medication to avoid polypharmacy and ‘dilution’ of positive effects of drugs under study.
Inclusion of organ-specific endpoints and disease activity measures.
Increase in number of adequately trained trial sites (recruitment, infrastructure and training).
Academia versus industry-driven clinical trials.
The committee wishes to acknowledge the support of the EULAR Standing Committee on Clinical Affairs. The committee also expresses its sincere appreciation and gratitude to the EULAR Secretariat and especially to Patrizia Jud, executive assistant, for the outstanding organisation.
GB and DTB contributed equally.
Handling editor David S Pisetsky
Contributors AF and MK performed the systematic literature review (SLR) and AF drafted the manuscript. GB supervised the methodology of the SLR and edited the manuscript. DTB convened and supervised the project and edited the manuscript. All authors edited the manuscript and accepted its final form.
Funding AF was supported by an Articulum Fellowship and a grant from the Hellenic Society of Rheumatology during the completion of this work.
Competing interests AF reports personal fees from GSK, Abbvie, Amgen, Enorasis and Genesis Pharma, outside the submitted work. MA reports fees from advisory boards from Novartis, Pfizer, Roche. IB reports personal fees from consultant for GSK, outside the submitted work. JNB reports grants from GSK, personal fees from GSK, personal fees from Abbvie, personal fees from UCB, personal fees from Enorasis, grants from Pfizer, outside the submitted work. RC reports personal fees from GSK, personal fees from Astra Zeneca, personal fees from Rubió, outside the submitted work. DJ reports personal fees from Astra-Zeneca, Aurinia, Boehringer-Ingeleheim, Celgene, BMS, Chemocentryx, grants and personal fees from GSK, from null, outside the submitted work. AK reports grants from Biogen, grants from Galderma, grants from GlaxoSmithKline, grants from Leo Pharma, personal fees from La Roche Posay, outside the submitted work. MM reports personal fees from GSK, Lilly and UCB. MS reports grants from GSK, UCB, Abbvie, outside the submitted work. JSS reports grants from AbbVie, Astra-Zeneca, Janssen, Lilly, MSD, Novartis, Pfizer and Roche, and personal fees from AbbVie, Amgen, Astra-Zeneca, Astro, BMS, Celgene, Celltrion, Chugai, Gilead, ILTOO, Janssen, Lilly, MSD, Novartis-Sandoz, Pfizer, Roche, Samsung, Sanofi and UCB, during the conduct of the study. AT reports personal fees from UCB, Pfizer, Abbvie, BMS, Sanofi, Roche, GSK, Alpha Sigma, Lilly, Jannsen, Cellgene and Novartis, outside the submitted work. RvV reports grants from BMS, GSK, Lilly, Pfizer, UCB Pharma, personal fees from AbbVie, AstraZeneca, Biotest, Celgene, GSK, Janssen, Lilly, Novartis, Pfizer, Servier, UCB, outside the submitted work. JW reports grants from GSK, grants from Incyte, personal fees from Biogen, personal fees from Leo, other from Novartis, during the conduct of the study. GB reports grants from GSK, Pfizer and personal fees from GSK, Abbvie, UCB and Enorasis, outside the submitted work. DTB reports unrestricted grant support/advisory board fees from Abbvie, BMS, Celgene, Enorasis, GSK, Pfizer, Novartis, UCB, Lilly, all deposited to the research account of the National and Kapodistrian University of Athens.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.