Bioethics Paper

The management of systemic lupus erythematosus (SLE) has evolved over the past few decades, leading to improved patient survival in the mid- term; how- ever, patients with SLE are still twice as likely to die as age- matched individuals in the general population and mostly die of long- term complications1–3.

Therapeutic strategies should aim to control disease activity while minimizing damage accrual related to both active disease and drug- related adverse effects4; however, the management of SLE is highly variable, possibly as a result of a lack of consensus on the con- cepts of remission and/or low disease activity (LDA) and on how SLE should be handled in the long term. These caveats could lead to suboptimal therapeutic strategies5. Nevertheless, scientific advances in diagnostics and dis- ease monitoring are encouraging the discussion of early diagnosis as well as personalized therapy. In this Review, we discuss the available and emerging therap eutic strategies in SLE and how such strategies can exploit currently available drugs to improve patient prognosis in the long term.

Importance of early diagnosis The median lag time from SLE onset to diagnosis has decreased from ~50 months before 1980 (ref.6) to 6–25 months since the year 2000 (refs7–9), largely because of the availability of antinuclear antibody (ANA) assays that enable an early diagnosis of SLE. However, this lag

time is unsatisfactory as autoimmune abnormalities can occur up to 10 years before clinical onset of SLE10,11. Furthermore, even when dealing with the interpreta- tion of well- characterized antibodies (for example, anti- double-stranded DNA (dsDNA) antibodies), caution is required as false- positive results can occur in cases of infection or malignancy and in the elderly population12; additionally, a diverse array of laboratory techniques are used for their determination13. Hence, the search for biomarkers of autoimmune abnormalities continues12.

Early diagnosis of SLE is important as it provides the chance of a timely treatment to improve the patient out- come and as the attainment of an early response damp- ens the disease course by minimizing organ damage14–16. Accordingly, in a retrospective longitudinal matched- cohort study of >9,000 patients with SLE, the frequency of flares was lower in the group of patients diagnosed within 6 months of symptom onset than in the group of patients with a later diagnosis, as was the rate of hos- pitalization and SLE- related costs9. Furthermore, data on lupus nephritis indicate that a delayed renal biopsy and consequently a delayed initiation of treatment are strongly associated with an increased risk of adverse renal outcomes, including renal failure and death17–21.

The measurement of potential biomarkers that change before clinical SLE, such as a rise in serum levels of pro- inflammatory mediators such as IL-5, IL-6, IFNγ and IFNα, has been limited to research settings22,23 and is

New therapeutic strategies in systemic lupus erythematosus management Mariele Gatto, Margherita Zen, Luca Iaccarino and Andrea Doria*

Abstract | The current treatment approach for systemic lupus erythematosus (SLE), as outlined in the recommendations by international medical associations including EUL AR and the ACR , is mostly eminence- based rather than evidence- based. However, knowledge on SLE is growing quickly , and such new advances need to be translated into clinical practice. Questions remain regarding the choice and timing of drug administration and tapering until withdrawal, which both can affect the balance between the control of disease activity and damage to organs triggered by long- standing and/or disproportionate immunosuppression. Currently , the treating physicians of patients with SLE are required to weigh the present with the future situation of their patients in an optimized balance between therapeutic harm and benefit. In this Review , the available therapeutic strategies and main challenges in the approach to SLE treatment are discussed. Remission and low disease activity are desirable therapeutic goals. Although the drug armamentarium for SLE has not expanded much in the past few decades, there are nonetheless opportunities to make better choices and explore combination therapies; such opportunities offer the potential of a personalized medicine strategy.

Unit of Rheumatology, Department of Medicine, University of Padova, Padova, Italy.

*e- mail: [email protected]

https://doi.org/10.1038/ s41584-018-0133-2

w w w.nature.com/nrrheum

R e v i e w s

30 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

not currently used to predict future disease development in asymptomatic individuals. Importantly, patients often present with fewer disease- specific elements the earlier they are diagnosed. Thus, a means of excluding diseases that mimic SLE (SLE mimickers)24 is paramount in the early stages of disease (Table 1).

Preventive strategies In addition to early diagnosis and treatment, preventive strategies should be adopted as early as possible for indi- viduals at risk of developing SLE (primary prevention) or for patients who have already been diagnosed with SLE to avoid disease exacerbations (secondary pre- vention) and disease progression (tertiary prevention) (fig. 1). The treatment of comorbidities is also paramount (box 1) as these conditions are the ultimate cause of death in a proportion of patients1.

Primary prevention Stratification of asymptomatic autoantibody- positive individuals according to other risk factors (such as whether patients have hypergammaglobulinaemia, reduced levels of C3 and/or C4 or a family history of SLE) might be useful for characterizing the likelihood of disease progression8,25. Indeed, although the presence of a low ANA titre on an isolated occasion might not require further investigation, a persistent high ANA titre (>1:80) and/or the presence of select autoantibodies (such as antibodies against dsDNA, U1RNP, ribosomal P or Sm) should be closely monitored, especially for patients who are at an increased risk of developing SLE (for example, pregnant women)26. Evidence of the usefulness of pri- mary preventive measures in asymptomatic individuals with serological abnormalities is mostly eminence- based; thus, the physician must base the decision on whether to implement such measures on patient- specific fea- tures. Similarly, whether or not to screen asymptomatic individuals who are potentially at risk of SLE is unclear27.

Preventive measures in SLE include the removal of modifiable risk factors (such as exaggerated sunlight exposure, smoking and drugs that can induce SLE)25,26; however, there remains a lack of consensus on what preventive pharmacological interventions can be used. Vitamin D supplementation might be advisable in

asymptomatic individuals to provide potential immuno- modulator y effects without notable drug adverse events26,28. Although ANA positivity alone might not necessitate hydroxychloroquine treatment, this therapy might be used for individuals who have a composite serology (such as positivity for anti- dsDNA or specific anti- extractable nuclear antigen (ENA) antibodies) and/or low complement levels because the risk of pro- gression is higher in these patients than in patients with ANA positivity alone25,28.

Another aspect of primary prevention concerns the risk of thromboembolic events in patients with SLE who are positive for antiphospholipid antibodies (aPLs) and have no history of thrombosis. Importantly, patients with SLE have an increased risk of thrombosis, compared with the general population26, that might be worsened by pro- thrombotic risk factors such as smoking, genetic hypercoagulability, renal disease or glucocorticoid use; these factors should be assessed at diagnosis and removed whenever possible. Second, asymptomatic aPL- positive individuals who are posi- tive for multiple aPL serological tests (double or triple positive) are in turn at increased thrombotic risk29,30 and might benefit from low- dose aspirin; this recom- mendation is supported by results from a meta- analysis and expert opinion31,32, although, conversely, a previous randomized controlled trial (RCT) has reported that this treatment provided no additional protection in asymp- tomatic aPL- positive individuals33. The occurrence of potentially precipitating conditions, such as pregnancy, prolonged immobilization or surgery, also requires, in our view, a temporary thrombosis prevention strategy, even in patients bearing a single yet persistent and high level of aPL specificity.

Secondary and tertiary prevention In SLE, the prevention of disease progression and flares is important to preserve organ function and avoid irre- versible damage (for example, to avoid the development of conditions such as end- stage renal disease associ- ated with persistently active lupus nephritis34, neuro- psychiatric sequelae, scarring alopecia or skin atrophy or dilatative myocardiopathy35).

Additionally, the occurrence of severe manifestations can worsen disease prognosis1 and increase disease- related medical costs36. In a 2018 observational study of 499 patients with lupus nephritis who were diagnosed between 1970 and 2016 (median follow- up 10.6 years (interquartile range (IQR) 4–18)), the patients had a milder presentation of lupus nephritis over time37. Interestingly, the time between SLE onset and the occurrence of lupus nephritis increased between 1970 and 2016 (from 1.3 ± 1.3 to 4.6 ± 6.3 years). The global improvement in presenting phenotypes might be the result of an earlier SLE diagnosis leading to a closer sur- veillance of patients and earlier and more appropriate therapeutic interventions, including the extensive use of antimalarial drugs, mycophenolate mofetil (MMF) and biologic drugs that can hinder the development of lupus nephritis. The question remains open as to whether more intensive immunosuppression at the time of SLE diagnosis might prevent the occurrence lupus nephritis.

Key points

• early diagnosis and early treatment are required for a better outcome in systemic lupus erythematosus (Sle).

• Preventive strategies should be applied at any stage of the disease course to minimize disease evolution or worsening; potential comorbidities should be prevented from the start of Sle treatment.

• The achievement of clinical remission and subsequent tapering of glucocorticoids until withdrawal are desirable subsequent steps in Sle management.

• even when remission cannot be attained, the treatment of patients with Sle should be optimized to achieve the lowest stable level of disease activity.

• Tapering of treatment should be initiated once there is a stable response and requires careful monitoring.

• Patient- tailored therapeutic strategies should consider the immunological background, clinical features, realistic potential for recovery and the expectations of each patient.

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 31

Treat- to-target approach In the wake of the treat- to-target (T2T) approach in rheumatoid arthritis (RA), remission and LDA have been proposed as targets for treatment in SLE and have been the subject of a number of studies38–48. The validity of a definition of remission and LDA should be on the basis of its ability to enable the identification of patients who achieve better outcomes39. Importantly, for patients who achieve remission or LDA, the target progresses to the maintenance of remission (such as the avoidance of disease flares, which should be the aim of maintenance treatment strategies).

Best outcomes of the T2T approach Remission. An agreement on what principles should guide the development of the definition of remis- sion in SLE was achieved in 2016 in the context of the Definition Of Remission In SLE (DORIS) project, which

involved a large international task force of 60 speci- alists and patient representatives who also put forward a definition of remission in SLE39. In addition to this proposed definition, at least three other different defi- nitions of remission have been proposed in the past few years39–41,43 (Table 2).

All four definitions distinguish two subtypes of remission — namely, complete (no serological or clin- ical activity) and clinical (clinically quiescent disease with serological activity permitted) remission. The defi- nitions differ in terms of the therapies allowed and the disease activity indices used (Table 2).

Compared with two decades ago, a greater propor- tion of patients now achieve remission49–52, and yet until a few years ago, prolonged remission was rarely reported51,52. However, a number of studies in the past few years have reported an increased prevalence of pro- longed remission44,46,47. This increase could be because

Table 1 | Common mimickers of SLE

Type Mimicker Similarity to SLE Differences from SLE

Rheumatologic– immunological conditions209

• Undifferentiated connective tissue diseases

• Undifferentiated polyarthritis • Undifferentiated spondyloarthritis • Antiphospholipid antibody syndrome • Connective tissue diseases (early- stage) • Vasculitis (early- stage) • Fibromyalgia • Sarcoidosis

• Generalized symptoms • Skin manifestations • Neurological manifestations • Thrombosis • Kidney involvement • ANA positivity

• Different autoantibody profiles • Additional clinical features and/or signs not

present in SLE (for example, Gottron papules in early stage of dermatomyositis)

• Histological analysis of skin or kidney samples might show lesions that are not typical for SLE (for example, acute thrombotic microangiopathy , intrarenal vascular lesions and/or arterial and arteriolar recanalizing thrombosis in antiphospholipid- mediated nephropathy210)

Other autoimmune diseases209

• Autoimmune hepatitis • Idiopathic thrombocytopenic purpura • Autoimmune haemolytic anaemia • TTP • MAS

• Organ- specific presentation might be clinically indistinguishable

• ANA positivity

• Additional clinical manifestations might help distinguish an SLE mimicker from SLE, but blood and immunological investigations are necessary

• Anti- dsDNA antibodies are normally absent or low in SLE mimickers

• Presence of schistocytes and anti- ADAMTS13 antibodies (in TTP)

• Typical histological pattern observed with piecemeal necrosis (in autoimmune hepatitis)

Infections24 • Viral infections (for example, parvovirus B19211, EBV212, CMV213 or HIV214 infection)

• Bacterial infections214 (for example, Treponema pallidum or Borrelia burgdorferi infection)

• Fungal infections (Trichophyton infection)215

• Parasitic infections (for example, Leishmania spp. or Toxoplasma spp. infection)216

• Lymphadenopathy • Blood cell count

abnormalities • Arthritis • Autoantibody positivity

(for example, seropositivity for ANA , anti- dsDNA anti bodies, anti- ENA antibodies and/or ACA)

• Hypocomplementaemia (occurs in >40% of individuals with a parvovirus B19 virus infection)

• Medical history (timing of symptoms and presence of risk factors for specific infections such as whether borreliosis is endemic in the area)

• High, rapid- onset fever with increased acute phase reactants (CRP more reliable than ESR) a feature of infections

• Hepatomegaly and splenomegaly more common with infections

• Severe SLE features (for example, lupus nephritis and NPSLE) and Raynaud phenomenon are unlikely features of infection

• Complement consumption is uncommon with infections

• Serological findings might help differentiate SLE from mimickers

Neoplasms24 • Malignancies217 (for example, Burkitt lymphoma, large B cell lymphoma, plasmacytoid dendritic cell neoplasm or T cell lymphoma)

• Benign conditions (for example, Kikuchi disease218 or Castleman disease219)

• Systemic symptoms and blood count alterations

• ANA positivity

• Specific manifestations of SLE and SLE- specific antibodies usually absent

• Acute phase reactants are likely increased

Medication- related or vaccine related24

• Medication (for example, IFNα or anti-TNF antibodies220)

• Vaccine (for example, HPV vaccine221)

• Shared clinical features (for example, skin manifestations and arthralgias)

• ANA positivity

• Medical history • Anti- histone antibodies more frequent in

drug-induced SLE

ACA , anti- cardiolipin antibodies; ANA , antinuclear antibody ; CMV, cytomegalovirus; CRP, C- reactive protein; dsDNA , double- stranded DNA ; EBV Epstein–Barr virus; ENA , extractable nuclear antigens; ESR , erythrocyte sedimentation rate; HPV, human papilloma virus; MAS, macrophage- activating syndrome; NPSLE, neuropsychiatric SLE; SLE, systemic lupus erythematosus; TTP, thrombotic thrombocytopenic purpura.

w w w.nature.com/nrrheum

R e v i e w s

32 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

of the application of new definitions of remission, together with improved knowledge and management of the disease.

The definitions of remission put forward by van Vollenhoven et  al.39 and Zen et  al.40 are similar, but unlike the definition by van Vollenhoven et al., the Zen et al. definition does not include the Physician Global Assessment (PGA), which has the known limitation of a relevant interobserver variability53. Moreover, pre- laboratory and post- laboratory PGA scores can differ54. Some researchers have pointed out that, despite these limitations, maintaining the inclusion of the PGA in the definition of remission could compensate for gaps in the SLE Disease Activity Index-2000 (SLEDAI-2K) (that is, the SLEDAI-2K does not consider haemo- lytic anaemia, myelitis and gastrointestinal activity)39.

However, it could be argued that the inclusion of a ‘treat- ment’ criterion in the definition of remission (namely, the threshold of prednisone equivalent of ≤5 mg/day and a stable dose of immunosuppressive drugs) in addi- tion to the SLEDAI-2K enables the exclusion of patients with active disease, even in the domains not covered by the SLEDAI-2K40.

Interestingly, comparable results in terms of the prevalence of remission and the protective effect of remission on damage progression can be achieved using either the definition by Z en et  al. 40,47 or van Vollenhoven et al.43,46, although it should be noted that these definitions were tested in different cohorts and were not tested side by side in the same cohort. Such find- ings suggest that an achie vement of a clinical SLEDAI-2K (that is, SLEDAI-2K score excluding serological

Subjects at risk of disease development • Serology • Abnormalities in the balance of pro-inflammatory and anti-inflammatory mediators

Prevention of disease developmenta • Smoking cessation • Vitamin D • HCQ • Asymptomatic aPL-positive individuals: low-dose aspirin31,32 • Tight follow-up

Prevention of disease flares Primary strategies in addition to control of disease activity using appropriate therapy68-70, 100 • Non-life-threatening: MTX, MMF, AZA or belimumab120,121,132 • Life-threatening: CYC or RTX

Delay of disease progression and/or damage • Primary and secondary strategies in addition to steroid-sparing strategies86-91

Fu n

ct io

n al

c ap

ac it

y Bu rd

e n

o f d

ise ase

Secondary prevention

Primary prevention

Tertiary prevention

Patients with ILE or SLE Diagnosis

Damage

Prevention and treatment of comorbidities

Primary prevention

Secondary prevention

Tertiary prevention

a

b

Fig. 1 | Levels of prevention in SLE. a | Therapeutic strategies in systemic lupus erythematosus (SLE) should aim to prevent disease development in predisposed individuals (primary prevention) and to prevent disease flares and progression in patients already affected (secondary and tertiary prevention). Secondary and tertiary preventive strategies also apply to those individuals who do not fit the SLE classification criteria but fulfil the diagnostic criteria for so- called ‘incomplete lupus erythematosus’ (ILE). The prevention of comorbidities should start at the time of diagnosis. Applying preventive strategies early during the disease course can help to avoid the establishment of organ damage, which is a major trigger of further damage and functional decline. b | Preventive measures, including pharmacological and behavioural strategies, can be implemented during the various stages of prevention. aPL , antiphospholipid antibody ; AZA , azathioprine; CYC, cyclophosphamide; HCQ, hydroxychloroquine; MMF, mycophenolate mofetil; MTX, methotrexate; RTX, rituximab. aThe advice for drug measures is potentially useful, but this advice is mostly eminence- based and strong evidence to support this advice is lacking.

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 33

activity) = 0 is probably the main driver of the protective effect of remission. Thus, the PGA is useful in grading clinical disease activity, but under conditions in which an instrument exists to determine the absence of activity (that is, clinical SLEDAI-2K = 0), studies to determine whether the PGA is redundant could be proposed.

All the newly reported definitions of remission (Table 2) have succeeded in identifying patients who achieved a better outcome in various studies, although differences in the design of the study, duration of follow- up and the type of cohort might explain some of the discrepancies between results.

A threshold for a durable remission has not yet been unanimously defined; however, the available studies suggest that the longer the remission, the better the protective effect against damage44,46,47. In this regard, a remission lasting 2 consecutive years proved to be the shortest duration associated with protection from damage in a cohort of 293 patients44. Importantly, the effect of long- standing glucocorticoid therapy, which is required to maintain clinical remission, should be considered. Indeed, in the long term, even a low daily prednisone dosage (≤5 mg/day) can contribute to damage accrual44. Thus, two major sequential steps in the SLE T2T approach can be identified: first, achieve clinical remission; and second, minimize or withdraw prednisone whenever possible (fig. 2).

Low disease activity. The concept of using LDA as a target has been applied to SLE in the past couple of years41–43, and preliminary data suggest that the achieve- ment of LDA is associated with better short- term out- comes (Table 3), although data on long- term outcomes are not available yet. Three definitions of LDA have been proposed (Tables 2,3).

Although no agreed- upon definition of LDA in SLE exists, an increasing number of studies have applied the definition by Franklyn et al.42, referred to as the lupus LDA state (LLDAS). The LLDAS was frequently attained in different cohorts14,42,45,47 with a high propor- tion of patients (ranging from 38.2%42 to 64.5%47) per- sisting in the LLDAS for ≥50% of the follow- up period. Importantly, a persistent LLDAS (that is, LLDAS in ≥50% of observations) is associated with a lower risk of damage accrual42,45,47, a finding also confirmed in a large cohort of 1,356 patients48, and a ≥2-consecutive- year LLDAS is an independent protective factor against new damage45. Conversely, failure to achieve an LLDAS at 6 months is an independent predictor of early damage15.

Notably, a similar protective effect on damage was detected if patients spent <25% of follow- up time in clin- ical remission, which means that remission is superior to LLDAS in preventing damage progression48.

The definition of LDA proposed by Polachek et al.41 is different from the two other new LDA definitions as it considers clinical SLEDAI-2K instead of SLEDAI-2K. The use of any medications for SLE, with the exception of antimalarial drugs, prevents the fulfilment of this defi- nition41. As such, this definition of LDA was associated with better disease outcomes after 2 years of follow- up. No external validation has been published.

In our opinion, LDA is primarily a clinical concept; therefore, clinical features should be considered over serological features for its definition, similar to what has been done for the definition of remission. In this regard, the LLDAS definition by Franklyn et al.42 has the limita- tion that SLEDAI-2K ≤4 and not clinical SLEDAI-2K ≤4 was used as an entry criterion, which means that both

Box 1 | Prevention of SLE comorbidities

The prevention of systemic lupus erythematosus (Sle) comorbidities should start from the time of diagnosis. Major complications are mostly related to long- standing immunosuppression and glucocorticoid therapy.

Cardiovascular disease • managing traditional risk factors200

- Control lipid levels in patients with dyslipidaemia with statins (or ezetimibe if there is a risk of myotoxicity)

- High blood pressure (<130/80 mm/Hg): angiotensin converting enzyme inhibitors indicated

- Smoking cessation • managing disease activity

- Hydroxychloroquine (protective against thrombosis and accelerated atherosclerosis)101

- Steroid- sparing strategies

Infections • Screening for latent infections (tuberculosis, hepatitis C virus, hepatitis B virus and HIv)

• Immunization with the inactivated influenza vaccine and the 23-valent pneumococcal polysaccharide vaccine (23-PPv) is strongly advised whereas immunization with the human papilloma virus (HPv) vaccine should be considered in young women with stable or inactive Sle201,202

• Prophylactic treatment with trimethoprim–sulfamethoxazole in patients with a low CD4 cell count (<200 cells/mm3)203

• Prophylactic treatment with a quinolone antibiotic (such as levofloxacin (500 mg, daily) or ciprofloxacin (500 mg, twice daily)) in patients with chronic neutropenia (<500 cells/mm3)a, potentially combined with antifungal therapy (refer to a specialist in infectious diseases)204

• modulation of immunosuppressive therapy

Cancer • most neoplasms that are more common in Sle than in the general population are

haematological (such as non- Hodgkin lymphoma), but solid neoplasms (such as lung, hepatocellular or cervical tumours) are also more common205,206

• Screening according to age- related and sex- related recommendations

• HPv vaccination and regular gynaecological screening in female patients, including the Papanicolaou (PAP) test (for women aged 21–30 years, repeated every 3 years) or the PAP test and the HPV test (for women aged 30–65 years, repeated every 5 years)

Osteoporosis207,208

• Non- pharmacological approaches - Reduce caffeine and alcohol intake - Smoking cessation - Weight- bearing exercise for at least 30 minutes daily - Adequate protein intake - Use of hip protectors in patients at risk of falls - Physical or occupational therapy

• Pharmacological approaches - vitamin D and calcium supplementation - Bisphosphonatesb or denosumab in patients at risk of fractures. Teriparatide if fracture occurs under bisphosphonate therapy

Glucocorticoid- induced osteoporosis • Bisphosphonates in patients receiving prednisone equivalent ≥7.5 mg/day for at least

3 monthsc

aThe National Comprehensive Cancer Network considers the cut- off for severe neutropenia to be 100 cells/mm3, but the risk of opportunistic infections is already high for patients with <500 cells/mm3; the appropriate course should therefore be judged by the physician. bDrug holidays are recommended. cNot if pregnancy is planned.

w w w.nature.com/nrrheum

R e v i e w s

34 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

anti- dsDNA antibody positivity and the presence of low complement serum levels would preclude attainment of LLDAS irrespective of the type of clinical manifestation present, even if scored as 1 or 2 by the SLEDAI-2K.

Notably, to capture low- intermediate disease activity, the measurement of disease activity incorporated into the LDA definition should be continuous and not categorical (that is, the presence or absence of an item; as it is in the SLEDAI-2K and SELENA- SLEDAI). In fact, LDA not only should correspond to milder lupus manifestations but also should identify individuals with LDA irrespec- tive of the type of manifestations (for example, low per- sistent proteinuria or mild arthritis). As a dichotomic

score, SLEDAI-2K does not capture the level of disease activity inside a given organ or domain (for example, the score does not change according to the number of joints involved or according to the severity of leukopenia or thrombocytopenia). In other words, SLEDAI-2K is able to discriminate between mild and severe lupus on the basis of the type of organ involvement but is unable to discriminate between low, moderate and severe disease activity inside a given domain. In this regard, the PGA, despite its limitations, could complement the SLEDAI.

The PGA threshold differs in the definition of remis- sion by van Vollenhoven et al.39 and the definition of LLDAS by Franklyn et al.42 (0.5 and ≤1.0, respectively);

Table 2 | Different definitions of remission and LDA

Study Name Indication of disease activity

Treatment Effect of remission or LDA status on damage, OR (95% CI)a

Clinical Sero- logical

PGA Prednisone (mg/day)

Antimalarial drugs (allowed)

Immuno- suppressive therapy (allowed)

Long- term (≥5 years) status

Other durations of status

Remission

van Vollenhoven et al.39 (DORIS)

Complete remission

No No <0.5 0 Yes No 2.42 (1.50–3.89)46,b

N/A

Clinical remission

No Yes <0.5 0 Yes No

Complete remission off treatment

No No <0.5 1–5 Yes Yes

Clinical remission on treatment

No Yes <0.5 1–5 Yes Yes

Zen et al.40 Complete remission

No No – No Yes No 2.52 (1.27–4.99)44,b 0.20 (0.07–0.53)47

0.228 (0.061–0.850) if in remission for 2 years44

Clinical remission off glucocorticoids

No Yes – No Yes Yes

Clinical remission on glucocorticoids

No Noc – 1–5 Yes Yes

Polachek et al.41 Remission No Yes – 0 Yes No N/A N/A

Ugarte- Gil et al. (GL ADEL)43

Remission off therapy

No No – 0 Yes No N/A 0.66 (0.48–0.9)43,d

Remission on therapy

No No – 1–5 Yes Yes

LDA

Polachek et al.41 LDA Noc,e Yes – 0 Yes No N/A N/A

Franklyn et al.42 Lupus LDA state

Noc,f Yes ≤1 ≤7.5 Yes Yes 0.071 (0.023– 0.217)46

0.47 (0.28–0.79) if ≥50% of follow- up spent in LDA42,g 0.279 (0.08–0.92) if in LDA for 2 years44

Ugarte- Gil et al. (GL ADEL)43

Lupus LDA status

Noc,h Yes – ≤7.5 Yes Yes N/A 0.66 (0.48–0.9)43,d

–, item not included in the definition; DORIS, Definition Of Remission In Systemic Lupus Erythematosus; GL ADEL , Gruppo Latino Americano de Estudio de Lupus; LDA , low disease activity ; N/A , not available; PGA , Physician Global Assessment; SELENA , Safety of Estrogens in Lupus National Assessment; SLEDAI-2K , Systemic Lupus Erythematosus Disease Activity Index-2000. aFor the effect of remission and LDA on damage, the odds ratio refers to that obtained from multivariate analyses. For the Polachek et al. definition, no multivariate analysis has been performed; in a univariate analysis, the mean damage accrual was lower in the combined group that included patients either in LDA or in remission than in the group of patients not achieving LDA or in the group of patients in remission41. bRisk of increase in damage if remission status (of any subtype) is not achieved. cAllowed but not a requirement. dThe effect of remission and LDA was evaluated as the cumulative time spent by all patients (as a whole, and not by each single patient) in one of the two statuses. eSLEDAI-2K ≤2, including only one clinical manifestation of rash, alopecia, mucosal ulcers, pleurisy , pericarditis, fever, thrombocytopenia or leukopenia. fSLEDAI-2K ≤4, with no activity in major organ systems and no haemolytic anaemia or gastrointestinal activity , no new features of lupus disease activity compared with the previous assessment and PGA ≤1 (scale 0–3). gMean (s.d.) duration of follow- up was 3.90 (2.0) years. hSELENA- SLEDAI ≤4.

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 35

however, the SLEDAI thresholds are not necessarily dif- ferent because a clinical SLEDAI of 0 in the definition of remission by van Vollenhoven et al.39 can coincide with an SLEDAI of ≤4 in the definition of LLDAS by Franklyn et al.42. Thus, remission and LDA should be taken as two different levels of remission rather than two conceptually different targets (that is, the ‘absence of disease activity’ versus the ‘persistence of a low level of disease activity’). However, if these two targets, currently named ‘remission’ and ‘LDA’, are defined using the same instruments (as they are for the van Vollenhoven et al.39 definition of remission and the Franklyn et al.42 defini- tion of LLDAS), they can be regarded as sequential steps of an improved treatment response.

What is still missing in the SLE armamentarium is a single, simple instrument for measuring disease activity that is able to clearly separate remission from LDA and from high disease activity on a continuum, similar to the Disease Activity Score 28 (DAS28) in RA.

Renal outcomes. According to the EULAR/European Renal Association–European Dialysis and Transplant Association (EULAR/ERA–EDTA) recommenda- tions for the management of lupus nephritis55, the goal of immunosuppressive treatment should be a complete renal response (CRR; that is, a urinary protein–creatinine ratio (UPCR) <0.5 and a normal or near- normal glomerular filtration rate (GFR) (±10%)) or, at least, a partial renal response (PRR; that is, a reduc- tion in UPCR ≥50% (subnephrotic) and a normal or near- normal GFR (±10%)). PRR has a worse prognosis than CRR56 but might be an acceptable outcome when all treatment options have been exhausted or cannot be

used because of a high risk of adverse effects. However, a PRR must be achieved preferably within the first 6 months of treatment and no later than 12 months after treatment initiation57.

The UPCR measures the amount of protein in urine at a definite time point in relation to urinary creati- nine and is commonly used as a suboptimal means of approximating 24-hour proteinuria in clinical studies58. In extension studies of the Euro- Lupus Nephritis Trial59, the threshold of proteinuria (with preserved renal func- tion) that can predict a favourable renal outcome was reported to be 0.7–0.8 g/day60,61; the assessment of renal outcome was not improved by the addition of other parameters such as serum creatinine or urinary sedi- ment60. This proteinuria threshold was associated with a good long- term outcome (defined as serum creatinine concentration ≤1 mg/dl at 7 years after entry into the trial)60. However, 7 years might not provide sufficient reassurance for the preservation of renal function in young patients. Moreover, some data have highlighted the detrimental contribution of silent active kidney lesions62, thus raising the question of whether treating to histological targets rather than to clinical remission would be valuable. However, treating to histological tar- gets would require repeated biopsies, the applicability of which is still debated62,63 and which are not currently recommended in the absence of clinical worsening.

Hence, it is reasonable to assume that patients with active lupus nephritis should be brought to the clinical target of proteinuria of ≤0.7 g/day, and treatment should preferably be started within 3–5 months of clinical onset (that is, within the ‘window of opportunity’), as longer intervals between clinical onset and treatment initiation

Third stepSecond stepFirst step

Reduce and stop immunosuppressants

Minimize and stop glucocorticoids

Clinical remission

Active SLE Treat to target

Reduce immunosuppressants

Measure disease activity every 1–3 months

Minimize glucocorticoids

Clinical LDA

Main target

Alternative target

Sustained clinical remission

Measure disease activity every 4–6 months

Sustained clinical LDA

Fig. 2 | Proposed treat- to-target algorithm in SLE. The main target in the treatment of systemic lupus erythematosus (SLE) should be the attainment and maintenance of clinical remission. If the main target cannot be achieved, clinical low disease activity (LDA ; without considering serology) could be a suitable alternative target. The first step (which is common for the two threads) is to treat the patient according to disease manifestations, following available guidelines and recommendations and taking into consideration comorbidities and other patient factors (see fig. 3). If the target is achieved, the second step would be to reduce and/or minimize and stop glucocorticoids. In patients in sustained clinical remission, the third and last step would be the de- escalation of immunosuppressive therapy , and, in selected cases, the complete discontinuation of immunosuppressants. In patients in sustained LDA , the third step would be to decrease immunosuppressive therapy.

w w w.nature.com/nrrheum

R e v i e w s

36 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

are associated with worse outcomes20. The evaluation of a renal response might require several months of follow- up to provide enough time for proteinuria normalization (for example, in proliferative lupus nephritis64,65); thus, an adequate follow- up period is important for making treatment decisions, which should be neither delayed nor premature66,67.

Best treatment for T2T approaches In SLE, treating ‘right the first time’ is an important con- cept; that is, an adequate intensity of treatment should be provided to control disease activity as early as possi- ble. In fact, evidence suggests that earlier achievement of remission is associated with a better long- term out- come and early responders are likely to maintain a good treatment response in the long term15,16. Unfortunately, a proportion of treated patients still fall outside the desired ‘control zone’, reaching a suboptimal level of disease con- trol that would theoretically require a stronger treatment to reach a desirable target.

Indications for what treatment to use are available among international recommendations55,68–70. In this section, we discuss strategies that make good use of the available drugs and aim to achieve treatment targets without exposing patients to overtreatment.

Glucocorticoids. Glucocorticoids have been a mainstay of treatment of patients with SLE for 60 years71, and a boost in glucocorticoid treatment is commonly used to treat small spikes of disease activity5 despite there being no evidence that boosting steroids provides any stable benefit in the long term72.

The adverse effects of glucocorticoids, such as the association of treatment with damage accrual, have been clearly established73,74, and immunological mech- anisms accounting at least in part for the occurrence of suboptimal response to glucocorticoids in SLE have been proposed75. Interestingly, a number of studies have reported no differences in the outcome of diverse SLE manifestations, including lupus nephritis, when omitting

high- dose glucocorticoids in induction or maintenance therapy76–79. Notably, although glucocorticoids are indi- cated for severe manifestations such as lupus nephritis and neuropsychiatric SLE (NPSLE), the duration and dosage of treatment that are used in clinical practice have not been established in an RCT80.

Data on lupus nephritis exemplify the unresolved issue of whether glucocorticoid minimization during induction or maintenance treatment is safe81. In several cohort studies76–78 and in one small RCT79, researchers compared the use of high versus low doses of gluco- corticoids (ranging from 3 g/day76 to <20 mg/day77) during induction treatment; both approaches had comparable rates of renal responses by the end of follow- up (that is, at 6 to 12 months), but low- dose glucocorticoid ther- apy was associated with a better safety profile78. Long- term data on the effects of a steroid- free maintenance regimen are available from the RITUXILUP (Trial of Rituximab and Mycophenolate Mofetil Without Oral Steroids for Lupus Nephritis) RCT, in which an induc- tion treatment (comprising rituximab 1 g and intrave- nous methylprednisolone 500 mg fortnightly) followed by a maintenance treatment that included MMF but not glucocorticoids resulted in 52% of patients in complete remission and 34% of patients in partial renal remission at 1 year82. At 5 years, more than 80% of patients were still in remission (complete or partial) and 77% were still glucocorticoid- free83.

The RituxiRescue study used a similar therapeu- tic protocol to that of the RITUXILUP trial (in terms of the rituximab and MMF treatment) but involved patients who were already on glucocorticoid mainte- nance and who did not uniformly receive intravenous methylprednisolone during their induction therapy. In this study, glucocorticoids were actively decreased over time according to clinical practice84. Notably, 76.3% of patients treated with this regimen reached a renal response (either complete or partial) at 5 years and 63.2% of individuals were off steroids or were receiving a minimal glucocorticoid daily dosage by this time85.

Table 3 | Minimum requirement for fulfilling definitions of remission and LDA

Definitions Disease activity Prednisone (mg/day)

Antimalarial drugs

Immunosuppressive drugs

Biologic drugs

SLEDAI PGA

Remission

van Vollenhoven et al.39

Clinical SLEDAI-2K = 0 <0.5 ≤5 Yes Yes Yes

Zen et al.40 Clinical SLEDAI-2K = 0 – ≤5 Yes Yes Yes

Polachek et al.41 Clinical SLEDAI-2K = 0 – 0 Yes No No

Ugarte- Gil et al. (GL ADEL)43

SELENA- SLEDAI = 0 – ≤5 Yes Yes –

LDA

Polachek et al.41 Clinical SLEDAI-2K ≤2 – No Yes No No

Franklyn et al.42 SLEDAI-2K ≤4 ≤1 ≤7.5 Yes Yes Yes

Ugarte- Gil et al. (GL ADEL)43

SELENA- SLEDAI ≤4 – ≤7.5 Yes Yes Yes

All items (disease activity , prednisone dose and other therapies) should be fulfilled at the same time. –, item not included and/or not mentioned in the definition; GL ADEL , Gruppo Latino Americano de Estudio de Lupus; LDA , low disease activity ; PGA , Physician Global Assessment; SELENA , Safety of Estrogens in Lupus National Assessment; SLEDAI, Systemic Lupus Erythematosus Disease Activity Index; SLEDAI-2K , SLEDAI-2000.

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 37

However, controlled studies of the feasibility of oral glucocorticoid- free regimens for patients with lupus nephritis or SLE are currently lacking. In light of the available86–91 and upcoming therapeutics92,93, an advisable approach to glucocorticoid tapering and decreasing the long- term glucocorticoid daily dosage is to use combina- tion therapy (for example, MMF and calcineurin inhib- itors in lupus nephritis94,95) or to add a biologic drug to standard- of-care therapy86,88,96,97.

Low- dose oral glucocorticoids can be as effective as higher dosages (≥30 mg/day) in treating systemic active extrarenal disease98, suggesting that the interven- ing dosage could be routinely lowered. Hence, during disease reactivation, glucocorticoids should be used at the lowest effective dosage and with a clear indication and duration of treatment. Physicians should perform a careful patient evaluation to ensure they are really treating the manifestations related to active disease. The dosage of gluco corticoids should be reassessed at every patient evaluation, which should fall at intervals of no longer than 4 months, even if the patient responds to treatment99. Finally, and most importantly, physicians should consider modifying the background treatment, especially when the patient has a relapsing–remitting disease course or is on long- term treatment with pre- dnisone ≥5 mg/day or an equivalent, and progressively abandon the tendency to correct disease activity by adjusting the glucocorticoid dosage alone.

Antimalarial drugs. Antimalarial drugs (for example, hydroxychloroquine and chloroquine) were initially recommended100 for the treatment of mild- to-moderate SLE manifestations, especially skin rashes and arthri- tis. In more recent studies, these drugs were shown to decrease disease activity and thrombophilia101 and improve patient survival102,103, and they are currently recommended for the treatment of active SLE68.

Two major aspects of antimalarial therapy have resulted in modifications to their recommended use. First, non- adherence to treatment can affect the disease outcome102,104, and non- adherence to antimalarial drugs might indicate poor adherence to all medication in gen- eral105. A number of factors can contribute to treatment non- adherence (including young age106,107), and non- adherence can lead to a worse disease course, misinter- pretation of disease activity and consequent unnecessary therapeutic changes105. Thus, counselling strategies are recommended for increasing patient compliance108, and blood hydroxychloroquine levels should be monitored in cases of adherence concerns107.

The second update on the use of antimalarial drugs concerns the new inclusion of hydroxychloroquine to the drugs recommended by the EULAR/ERA–EDTA international task force for the management of lupus nephritis55. Indeed, data from previous studies suggested that patients with lupus nephritis already being treated with hydroxychloroquine had a reduced likelihood of renal damage109,110 compared with those patients not receiving hydroxychloroquine treatment. Interestingly, some evidence suggests that administration of hydroxy- chloroquine in lupus nephritis is inversely associated with tubulo- interstitial inflammation111 and therefore

is likely also inversely associated with a reduction in interstitial fibrosis.

In summar y, antimalarial drugs have gained increased attention for the treatment of mild as well as severe SLE manifestations and should be regularly advised. Monitoring of hydroxychloroquine serum lev- els is not a common practice but might be considered in cases in which poor adherence is suspected before any therapeutic modifications are performed.

Immunosuppressive drugs. The traditional immuno- suppressive drugs used for the treatment of SLE include alkylating agents (cyclophosphamide), inosine monophosphate dehydrogenase (IMPDH) inhibitors (MMF and mycophenolic acid), selective inhibitors of purine and/or pyrimidine synthesis (azathioprine and methotrexate, respectively) and calcineurin inhibitors (cyclosporine and tacrolimus)112. Although these drugs do not target specific molecules and non- selectively sup- press various cellular processes, lymphocytes (excluding plasma cells) are the major cell type affected as these cells are highly proliferative and preferentially use metabolic pathways that are targeted by immunosuppressive drugs such as IMPDH inhibitors113.

The use of single immunosuppressive drugs is discussed in international recommendations else- where55,68–70,100. In this section, we discuss potential com- bination therapies. The purpose of combining different therapeutics is to exploit different mechanisms of action of the drugs and to lower the effective dosage of either drug. Interestingly, the effectiveness of combination therapies is usually superior to that expected from the sum of the single drugs, suggesting a synergy.

The effectiveness of combination therapies (such as combining a calcineurin inhibitor with MMF either as an induction or maintenance treatment) in SLE, and especially in lupus nephritis, has been explored in vari- ous studies91,114. Most data regarding the combination of tacrolimus and MMF in lupus nephritis are limited to its effectiveness during 6-month induction therapy, for which renal remission was attained more quickly and by a greater proportion of patients receiving tacrolimus– MMF combination therapy than those receiving intrave- nous cyclophosphamide (46% versus 26% at 6 months)90. In an extension study, patients who achieved renal remission with either regimen in the induction phase were monitored for an additional 18 months91; in the maintenance phase, the patients who had received com- bination therapy of MMF and tacrolimus during the induction phase continued with this treatment, whereas patients who had received intravenous cyclophospha- mide were switched to azathioprine. Both groups had similar rates of renal relapse by the end of follow- up91, in keeping with earlier data114. However, combination therapy was associated with a decreased incidence of adverse events compared with azathioprine alone, espe- cially in terms of leukopenia and liver dysfunction91. Conversely, the induction trial showed an increased rate of adverse events in the group of patients receiving combination therapy (namely, serious infections and varicella zoster reactivation90), thus raising questions on the actual safety profile of this combination therapy.

w w w.nature.com/nrrheum

R e v i e w s

38 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

The apparent discrepancies between the two studies might result from the higher dosages required dur- ing the induction phase than during the maintenance phase; however, as the pharmacokinetics of tacrolimus has a high interindividual variability, the plasma levels of this therapy should be carefully monitored and usually maintained around 4–6 ng/ml in SLE114.

The efficacy of low- dose or high- dose voclosporin (a calcineurin inhibitor) in combination with MMF for the treatment of active lupus nephritis has been compared with that of MMF alone115. The primary end point (com- plete response) was defined as preserved efficacy (UPCR ≤0.5 mg/mmol and estimated GFR ≥60 ml/min/1.73 m2 without a decrease ≥20% from baseline) together with a steroid- sparing effect and no administration of rescue medications. Notably, the rate of complete (and par- tial) remission was higher in the voclosporin group at 6 months and maintained higher at 12 months than in the group of patients receiving MMF alone115. However, it should be mentioned that some drug safety concerns have been raised following the higher number of deaths occurring in the voclosporin group (ten in the low- dose group), which is expected to be addressed in the ongoing phase III trial of this treatment in patients with lupus nephritis93.

Synergy between calcineurin inhibition and MMF in lupus nephritis has also been documented at the transcriptional level in samples of human renal biop- sies and in a mouse model of lupus- like nephritis; for example, the expression of some pro- inflammatory genes was inhibited following treatment with both drugs but not following treatment with either drug alone116. These observations, together with the finding that tacrolimus treatment in patients with lupus nephri- tis results in a lowered activity index after 6 months (as assessed by repeated renal biopsies)90, implicate the anti- inflammatory potential of calcineurin inhibition in lupus nephritis, despite existing concerns around its mechanism of action117.

Biologic drugs. Although a number of promising bio- logic therapies are in development92,118,119, those currently available and most often used in clinical SLE manage- ment are belimumab (an anti- B lymphocyte stimulator (BLyS, also known as TNFSF13B) antibody, which is thus far the only licensed biologic drug for SLE120,121) and rituximab (an anti- CD20 antibody).

Despite the failure of rituximab treatment in SLE RCTs122,123, rituximab is prescribed in clinical prac- tice124–126 and is included in the European and US rec- ommendations for the treatment of refractory lupus nephritis55,127 as well as NPSLE69. Moreover, on the basis of expert opinion, rituximab has also been suggested for the treatment of refractory and glucocorticoid- dependent disease with severe manifestations, includ- ing severe thrombocytopenia128 and recalcitrant skin disease129, although chronic lesions are unlikely to respond129.

Overall, the effect of improved adherence to standard- of-care therapy following enrolment into a study or the effect of commencing standard- of-care ther- apy for treatment- naive patients (in both the placebo and

the rituximab group) is speculated to have masked any marginal efficacy rituximab had in past RCTs80,130. Most investigations of rituximab administration in SLE before the commencement of immunosuppressive drugs have been carried out in patients with lupus nephritis82,84,97 and seldom in patients with non- renal SLE96,131. In such studies, rituximab consistently provided a relevant steroid- sparing effect83–86,96, whereas the ability of ritux- imab to maintain the clinical response varied among different cohorts84,96,131.

Belimumab can be efficacious in real- life settings, with increasing evidence suggesting that this bio- logic drug (as either an intravenous or subcutaneous formulation) can reduce disease activity and prevent flares86,87,91,132,133. Importantly, damage accrual was hin- dered after belimumab administration, with no sub- stantial change in the SLICC/ACR Damage Index (SDI) at 2 years86,89. As these results come from uncontrolled studies, caution is needed in interpreting these find- ings; however, the consistency of these findings across different cohorts86,89 makes this observation promising. Belimumab is currently being tested for the treatment of lupus nephritis134; notably, the use of belimumab resulted in a long- term reduction in proteinuria both in a trial extension135 and in real- life observational studies86,88.

Researchers have tested the coordinated strategy of using rituximab followed by belimumab in the treat- ment of refractory SLE136–138. The rationale for this approach resides in the peak in serum BLyS levels that follow B cell depletion139. An ongoing phase III RCT is currently testing this approach for the treatment of non- renal SLE140.

Drug costs and cost- effectiveness might limit the accessibility that patients have to biologic therapies. Indeed, studies of economic simulation models sug- gest that belimumab treatment could be cost- effective in several countries in Europe, including Italy, Spain, Portugal and Greece141–143, mostly because of the gain in life- years and quality of life and the abatement of indirect costs associated with treatment141. Conversely, the local costs of this treatment might hamper its cost- effectiveness in countries such as the United Kingdom, Germany and France144, thus increasing the disparities in drug utilization. Indeed, several unapproved treatments often might fail for a patient before they have access to belimumab80, likely because the incremental benefit of belimumab is not perceived as being large enough.

Deciding what timing and reasons are necessary for a patient to commence biologic therapy (that is, biologic accessibility) remains a relevant point as patients often must first demonstrate refractoriness to standard- of-care therapy including a traditional immunosuppressant before they can commence treatment with a biologic drug, limiting the accessibility of patients to belimumab in the early stages of disease, despite it being the only drug approved for SLE. In this regard, it should be mentioned that in recent RCTs of biologic therapies for the treatment of non- renal SLE92,120,121, eligible patients often did not need to have failed a previous immunosuppressive drug. Accordingly, up to 50% of patients enrolled in two RCTs of belimumab (BLISS-52 and BLISS-76)120,121 were not taking immunosuppressants, and consistently official indications

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 39

for belimumab do not necessarily require the failure of previous immunosuppressive drugs before use145.

Hence, for moderate SLE, glucocorticoids and anti- malarial drugs might be considered as the standard- of-care therapy, on top of which additional therapies are advised55,68–70. For patients with severe manifestations of SLE that are refractory to cyclophosphamide or MMF, biologic drugs (namely, rituximab) can also be advo- cated as, although B cell- depleting therapies can take longer to exert their therapeutic effects than alkylating or antimetabolite agents, B cell depletion might still pro- vide marginal benefits for refractory manifestations130. Finally, the prompt use of belimumab is advisable, at least in countries where this treatment is cost- effective, for the treatment of SLE and especially for patients with active disease and non- life-threatening manifestations146.

Despite the limited evidence available supporting the use of biologic therapies in SLE, maximizing the benefits of their use is paramount, taking into account the accessibility of these drugs and the additional ben- efit they can provide for patients with severe as well as refractory SLE. Indeed, it is common for patients to have residual disease activity (such as inflammatory arthral- gias with prolonged stiffness or suboptimal control of proteinuria after a nephritis flare) despite receiving full standard treatment, and such patients are likely to benefit from additional interventions to help stabilize the treatment response while rendering glucocorticoid tapering feasible.

Critical aspects in T2T trial design A suitable therapeutic strategy that clarifies which med- ications should be given or switched if the treatment goal is not reached has not yet been defined in SLE. This issue could be investigated in dedicated RCTs of T2T strategies, but such trials are in turn difficult to set up because of the heterogeneity in patient management. Several issues would arise when setting up a T2T trial, among which the method of patient selection, the type of treatment assessed and the included outcome meas- ures seem the most critical. Such aspects have been extensively debated in the field and are deemed at least partially responsible for the failure of diverse SLE trials80.

Advances in the management of lupus nephritis might inform the design of T2T strategies for SLE. In fact, a strategy for reaching the desired target (that is, CRR or PRR) in patients with lupus nephritis has already been proposed in the 2012 EULAR/ERA–EDTA recommendations55, which advise to change the type of induction therapy treatment in the absence of any improvement after 3–4 months or in the absence of (at least) a partial response after 6–12 months or complete response after 24 months. In patients with lupus nephri- tis for whom treatment with MMF or cyclophosphamide failed either because of lack of efficacy or adverse events, the treatment should be switched from MMF to cyclo- phosphamide, or from cyclophosphamide to MMF, or rituximab should be given55. The agreement on an over- arching strategy is of relevance as thus far no therapeutic approach is universally accepted to reach remission or LDA in non- renal SLE. Moreover, the EULAR/ERA– EDTA recommendations include set time points for the

evaluation of the renal response (6 and 12 months)55,57 and, most importantly, for the theoretical amount of time needed before tapering of maintenance therapy can be considered (3 years)147,148. Hence, in lupus nephritis, some key aspects have been clearly defined, including quantifiable targets, drugs to be used, how to modify the therapeutic approach in cases of a lack of response at definite time points and how long (at least) to continue maintenance therapy before considering tapering.

Transferring the procedure for lupus nephritis to non- renal SLE might be tricky for several reasons. First, remission and LDA, unlike renal response, are compos- ite targets that incorporate aspects of disease activity, therapy and physician assessment. In other words, these targets are a qualitative rather than a quantitative meas- ure of improvement. Second, improvements in organ- specific targets (for example, resolution of arthritis or skin rash) might not guarantee the attainment of sys- temic remission. Third, even when remission or LDA is achieved, its maintenance should be evaluated over a long time period (for example, at least 2 years is needed to evaluate whether damage progression has been hin- dered42,44), thus rendering these targets unsuitable for an RCT. Given these caveats, a number of aspects need to be addressed before proposing a T2T trial in non- renal SLE, including the choices of which patients to recruit, when to periodically assess the patients and how to assess the patients.

The method of patient recruitment could differ according to what aspect of T2T is under investigation; for example, whether the aim is to assess patients with active disease who need to reach remission or patients who are in remission and need to maintain the treat- ment target. Medication- naive patients with new- onset SLE are the best patients to recruit for trials assessing the best strategy for reaching remission. Conversely, for trials evaluating maintenance strategies, for which the primary outcome would be relapse, patients in remis- sion are the appropriate choice; such trials would be helpful for determining long- term therapeutic strat- egies. The two different trial types cover two rele- vant aspects of T2T (that is, how to reach the target and how to maintain it), the latter of which might be a major problem in the face of a relapsing–remitting disease course, a common feature of SLE149,150. Other issues relating to the selection of patients in T2T trials include which type and severity of the clinical pheno- type to include or exclude, which requires a dedicated discussion outside the topic of this Review.

To adequately evaluate the treatment response of patients, a suitable frequency of reassessment is neces- sary. Notably, for RA, the tight control of disease activ- ity through regular assessment (every 1–3 months) can result in a notably higher treatment response rate than a lower frequency of visits (every 6–12 months)151. In SLE, a similar strategy for tightly controlling disease activity might be suitable, as indirectly suggested in one study that compared the outcomes of patients with SLE being treated in community clinics with those being treated in a specialty clinic152. In this study, the patients with SLE who attended a community clinic had, on average, a higher level of disease activity, a higher cumulative

w w w.nature.com/nrrheum

R e v i e w s

40 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

glucocorticoid dosage and a lower hydroxychloro- quine intake at 5 years than the patients who attended a specialty clinic, suggesting that an expert- based man- agement of SLE results in better outcomes in terms of disease control and treatment handling. Although not directly compared, the factors responsible for this dif- ference are likely a higher frequency of visits (every 2–6 months) and more consistent assessment of disease activity, damage and comorbidities in the specialized care setting compared with the community care setting.

The method of patient assessment is an impor- tant consideration for T2T trials; ascertaining partial improvements in disease activity, including improve- ments not captured by current global disease activity scores, is necessary as such changes might be clinically relevant and affect therapeutic decisions. Further unre- solved issues include defining the standard- of-care therapy on which the additional therapies are tested and the type of intervention that should be given when the pre- specified treatment target is missed within a given time frame, which would probably require a switch to a different drug.

Once the issues highlighted in this section have been resolved, trials can be performed that compare differ- ent T2T strategies. The design of RCTs will need to be further tailored to different patient subsets (such as subsets with different levels of disease activity, types of disease manifestations, ethnicities and/or other clinical or immunological features).

Treatment tapering and withdrawal The recommended treatment approach for patients with inactive disease, particularly for patients in prolonged remission, is a matter of debate (fig. 3). The benefits of drug tapering and discontinuation have been proved for glucocorticoids44,153, but insufficient evidence is availa- ble regarding the tapering and eventual withdrawal of antimalarial drugs and immunosuppressive drugs. As a general consideration, during the tapering and withdrawal of treatment, close surveillance is necessary to detect early signs or symptoms of disease relapse, and the patients should be informed that, although they have inactive disease, routine laboratory tests and regular clinical evaluations are still needed.

Glucocorticoids SLE maintenance treatment should include the lowest glucocorticoid dosage needed to control the disease, and glucocorticoids should be completely withdrawn whenever possible4; however, further studies are needed to clarify when complete withdrawal of glucocorticoids can be done safely.

In a retrospective study of patients with SLE who were receiving 5 mg/day of prednisone, the prednisone dose could successfully be tapered to <5 mg/day for at least 1 year in 377 out of 688 patients (55%)154. Successful tapering was most common in patients with a fairly low level of disease activity, in patients without cutaneous or musculoskeletal activity or in patients diagnosed after the year 2000. Some data are available on the effects of glucocorticoid tapering in lupus nephritis, as dis- cussed in a previous section; in this subset of patients,

researchers have even attempted to decrease glucocor- ticoid dosage during the induction treatment (and thus, before the achievement of a renal response).

Whether the complete withdrawal of glucocorti- coids requires the achievement of complete remission or whether clinical remission or LDA can be sufficient has not been elucidated, and no trial has specifically addressed this issue. Frequently, patients who have been receiving long- term glucocorticoid therapy experience subjective difficulties following the withdrawal of gluco- corticoids, despite persistent SLE inactivity; in these cases, an ‘alternate- day’ glucocorticoid regimen or a slower taper might be feasible155.

Antimalarial drugs Few data are available on the contribution of antima- larial drugs to the maintenance of remission, and the actual protective effect of antimalarial drugs against SLE flares is not clear156–159. The only RCT on the effect of tapering hydroxychloroquine was conducted in 1991 by the Canadian Hydroxychloroquine Study Group156. The discontinuation of hydroxychloroquine in patients with stable disease was associated with a considerable increase in the risk of new clinical manifestations, exac- erbation of previous manifestations and a 2.5-fold higher relative risk of flare. Studies of adherence to antimalarial drugs can provide indirect evidence of the effects of anti- malarial withdrawal, as non- adherence might resemble an unplanned drug withdrawal.

In one study investigating the relationship between serum hydroxychloroquine levels and SLE activity, a higher serum concentration of hydroxychloroquine was associated with lower disease activity at baseline and a reduced likelihood of disease flares during follow- up104. In a longitudinal subgroup analysis of 73 patients in remission, there was a trend towards having lower levels of disease activity and number of flares over time for the group of patients who had therapeutic hydroxychloro- quine levels ( > 500 ng/ml) at baseline than for the groups of patients with lower baseline hydroxychloroquine lev- els, although the difference between the groups did not reach statistical significance160. Similarly, in a 7-month multicentre RCT of 171 patients with SLE who were ran- domly allocated either to stay on stable pre- study dose of 100–750 ng/ml hydroxychloroquine or to increase the hydroxychloroquine dose to achieve a target blood level of >1,000 ng/ml, the rate of SLE flares was not notably different between the two groups; nonetheless, it should be pointed out that an increase in hydroxychloroquine blood levels in the control group was detected, meaning that patient adherence improved after inclusion in the study, and this better compliance might have biased the results of the study159. Regarding lupus nephritis, maintenance of a hydroxychloroquine serum concentra- tion of ≥600 ng/ml is suggested to be protective against renal flares161; however, this notion is on the basis of retrospective observations.

Additional aspects should be considered when eval- uating the long- term use of antimalarial drugs in SLE remission, including that these drugs have favour- able lipid- lowering and glucose- lowering effects101, might contribute to the prevention of thrombotic and

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 41

thromboembolic events101, are safe and effective during pregnancy162 and might contribute to the preservation of renal function110.

Thus, taking into consideration the cumulative ben- eficial effects of antimalarial drugs, clinical experience suggests that their use is a safe therapeutic and preventive strategy that should be considered for all patients in remis- sion and that could be continued as long- term therapy. Nevertheless, for patients in complete, durable remission, the withdrawal of antimalarial drugs could be feasible, although no studies on the effects of antimalarial with- drawal are available to date. The decision of whether to discontinue antimalarial drugs during clinical remission should be on the basis of the patient clinical history, the presence of any comorbidities, the existence of damage, the duration of remission and patient preference.

Immunosuppressive drugs Few data are available on the effects of withdrawing immunosuppressive drugs from patients with SLE who are in remission, especially for patients with non- renal SLE. In one study, 156 out of 667 patients with non- renal SLE achieved at least one period lasting ≥1 year of drug- free clinical remission153. However, different results were obtained in a small controlled trial of azathioprine withdrawal in nine patients with stable non- renal SLE. In this study, disease flares occurred in seven patients after a mean interval of 10.5 weeks after drug withdrawal163. Furthermore, data from the Toronto Lupus Cohort showed that immunosuppressive drug- free prolonged remission (>2 years) was rare164.

Nevertheless, the results of these studies should be critically considered given the subsequent improvements

Active SLE

Discontinue corticosteroids

• Maintain or restore corticosteroids at the lowest effective dose • Introduce or change the immunosuppressants or biologics

Sustained clinical remission?

Sustained clinical remission?

Sustained clinical remission?

Reduce immunosuppression

Discontinuation of immunosuppressants

Follow-up screening for comorbidities and SLE activity

Continue antimalarials Sustained complete remission? (negativization of anti-dsDNA antibodies and normalization of complement levels)

Treat according to the disease manifestations and the patient's features (such as age, gender and the presence of comorbidities)

• Antimalarials (if not already prescribed) • Non-pharmacological interventions (smoking cessation and sun protection) • Minimize corticosteroids to the lowest effective dose

Clinical remission achieved?

Sustained clinical remission?

Yes

Yes

Yes

Yes

Yes

Yes

Consider stopping antimalarials

No

No

No

No

No

No

Fig. 3 | Proposed algorithm for the management of patients with SLE in remission. Presented is one approach that could be used for the management of patients with systemic lupus erythematosus (SLE) in remission. The first step in the management of patients with SLE who are in remission is to evaluate whether the patient has obtained the easier-to-achieve type of remission (that is, clinical remission rather than complete remission). If the patient is in clinical remission, a progressive reduction in therapy should be carefully undertaken, maintaining regular follow- up. Eventually , discontinuation of therapy might be considered. The term ‘sustained remission’ refers to a durable state of remission, which ideally should be associated with a lower likelihood of flare. Any clinical decision should be based on patient characteristics and the clinical observations of the treating physician. Notably , the time needed to achieve remission is a relevant prognostic factor as the shorter the duration of active disease (meaning a rapid achievement of remission), the easier it is to maintain the remission status. dsDNA , double- stranded DNA.

w w w.nature.com/nrrheum

R e v i e w s

42 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

that have been made in strategies for SLE management, including attempts at diagnosing SLE earlier, develop- ments in the tailoring of treatments to different man- ifestations and the availability of evidence- based and expert- based recommendations; thus, the scenario depicted by the aforementioned studies might be out- dated. Further studies aimed at identifying predictive factors of disease quiescence after drug discontinuation (for example, the type and duration of remission, the manifestations that require immunosuppressive therapy and the pattern of disease activity) would be timely.

Some data are available on the effect of immuno- suppressive drug discontinuation in lupus nephritis. In a study of 11 patients with class IV lupus nephritis165, the withdrawal of cyclophosphamide 2 years after com- plete renal remission resulted in a relapse in 36% of patients. In another study, 15 out of 33 patients (45%) with class IV lupus nephritis who withdrew immuno- suppressive therapy after treatment with intravenous cyclophosphamide and glucocorticoids experienced a renal flare166.

Researchers have tested a progressive approach of dis- continuation of therapy in 52 patients with lupus nephri- tis in durable renal remission. Immunosuppressive drugs were first de- escalated and withdrawn, and gluco- corticoid tapering was subsequently started: 32 out of 52  patients did not experience any flare during a median follow- up time of 101.8 months after drug dis- continuation. The patients who did not flare had a longer duration of treatment and remission before withdrawal and were concomitantly treated with chloroquine. Ten patients who experience a flare after the first drug withdrawal could later withdraw therapies and were free of immunosuppressive and glucocorticoid therapy after a median follow- up time of 286 months167. Importantly, the researchers recommended a tight follow- up dur- ing therapy de- escalation and an even closer follow- up after complete withdrawal of immunosuppressive agents and glucocorticoids (that is, every 15 days for the first 2 months, then every month for 6 months and then every 2–3 months).

Thus, immunosuppressive drug withdrawal requires careful evaluation tailored to each patient, taking into consideration the medical history, type and severity of organ involvement, the response to therapy (in particu- lar, whether remission was rapidly achieved or not), the number of previous relapses, the number of immuno- suppressive drugs and the duration of immunosuppres- sion needed to achieve remission. A reasonable duration of remission44 is required before considering drug dis- continuation, and progressive dose reduction, together with a close follow- up, could be a good option. Patients should be informed of the importance of continuing a regular follow- up after complete drug withdrawal.

Personalized medicine Personalized (or precision) medicine broadly involves tailoring therapy to the clinical features of a given patient to provide a customized approach for improving their disease; given the developments in omics techniques over the years, such an approach could ideally be used to tailor treatment to the underlying molecular pathways168.

Clinical predictors of drug response and effective methods of patient stratification are needed in SLE169,170, as demonstrated by the differences in drug responses of patients with diverging clinical features. For example, although a trial of epratuzumab (an anti- CD22 anti- body) for the treatment of patients with SLE failed, post hoc analysis indicated that those patients with both SLE and Sjögren syndrome had a better BILAG (British Isles Lupus Assessment Group) and BICLA (BILAG- Based Composite Lupus Assessment) response over time than the patients with SLE without secondary Sjögren syn- drome171. Furthermore, two phase II trials that stratified patients with SLE according to their manifestations had promising results118,119.

One of the latest aspects of personalized medicine concerns the incorporation of omics data (such as genomic, transcriptomic, proteomic and metabolomic data). Such integrated information can provide insights into the mechanisms of disease, which can lead to spe- cific treatments172. Indeed, using omics data, autoim- mune diseases can be dissected on the basis of molecular patterns and their resulting clinical manifestations, thus potentially providing a means of upstream stratification to indicate the most appropriate therapeutic approach. In 2017, a large omics- based project was initiated (PRECISESADS) to define a new molecular- based clas- sification of systemic autoimmune diseases on the basis of shared composite biomarkers172.

Several levels of stratification can be used in SLE to take into account susceptibility genes173,174, epigenomic data175,176, transcriptomic data173,177 and proteomic data178; however, limited metabolomic data are avail- able for SLE172. The genetic variability in SLE is large and gene microarray studies have implicated a num- ber of therapeutic targets in SLE179; however, the global clinical applicability of genetic stratification is still limited180. Epigenetic mechanisms involved in the reg- ulation of gene expression are greatly altered in SLE181, accounting for the detrimental yet unclear role of envi- ronmental factors in SLE. In patients with SLE, several genetic regions are often extensively hypomethylated, which results in the overexpression of normally regu- lated genes, particularly genes involved in interferon pathways182 and B cell activation181. Moreover, abnor- malities in histone acetylation patterns are associated with disease development via the upregulated expres- sion of survival factors and interferon regulatory fac- tors in (mostly) T cells and monocytes181. Interestingly, the presence of some epigenetic abnormalities, such as reduced activity of the epigenetic regulator miR-146a or hypomethylation of the gene encoding interferon regulatory factor 7 (IRF7), correspond with increased disease activity or specific manifestations183,184. Notably, commonly available drugs such as antimalarial drugs or the immunosuppressive drug mycophenolic acid can modulate epigenetic mechanisms in  vitro (for example, by affecting microRNA expression or his- tone acetylation patterns)185,186, and treatment of mice with the epigenetic modifier trichostatin A (a histone deacetylase inhibitor) can result in increased regulatory T cell numbers, modulating autoimmunity in a mouse lupus model187. These results suggest that modulating

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 43

mechanisms that alter gene expression might be a feasible approach.

Most evidence regarding the stratification of patients with SLE concerns grouping patients according to their interferon signature (although this signature is not only present in SLE but also in other autoimmune dis- eases)188. For example, T cells in patients with SLE have an abnormal expression of interferon- related genes189. In 2016, the first longitudinal analysis of the blood tran- scriptomic profiles of patients with SLE was performed: 158 paediatric patients with SLE were stratified according to their transcriptional profile, and the analysis identi- fied 5 immune signatures and 7 subgroups of patients190. Among the five immune signatures identified, the pres- ence of an interferon signature or a neutrophil signature (that is, the expression of neutrophil- specific transcripts in the blood or a given tissue) strongly correlated with disease activity and the development of lupus nephritis. Notably, an enrichment of neutrophil transcripts was detected during progression to active nephritis, and both the interferon and neutrophil signatures could be reduced with MMF treatment, especially in patients with proliferative lupus nephritis.

Notably, researchers have reported differences in the amount and type of transcripts that are upregulated in the affected glomeruli of patients with lupus nephritis, which might have a wide intersubject and intrasubject variability. Such heterogeneity suggests that any poten- tial personalized treatment strategy will have to take into account this high complexity that requires further study191. In the past few years, T cell abnormalities such as altered gene expression of signalling lymphocyte activation molecule receptors192 or enrichment of uri- nary CD4 cells193 have been identified in patients with lupus nephritis, which might be associated with a more refractory disease phenotype192. Interestingly, increased expression of neutrophil- related genes in the blood of patients with SLE is associated with renal inflamma- tion194,195, in keeping with previous findings suggesting the involvement of neutrophil- driven processes in the development of lupus nephritis174.

In the past couple of years, the stratification of patients with SLE has been attempted in clinical trials of type I interferon inhibition, in which patients were grouped according to the level of interferon- dependent genes92,196. Notably, in the genomic and proteomic stud- ies that identified IFNα as a potential therapeutic tar- get, other potential therapeutic biomarkers were also identified, such as BLyS, TWEAK (TNF- related weak inducer of apoptosis) and APRIL (a proliferation- inducing ligand)178. Despite the failure of treatment with atacicept, a drug that targets both BLyS and APRIL, in patients with SLE197,198, a subgroup analysis performed

on data from the ADDRESS II trial showed that higher baseline levels of BLyS or APRIL are associated with a greater treatment response to atacicept (as assessed by a decrease in flare rate) at 52 weeks199, thus re- enforcing the importance of patient subgrouping in the design of RCTs and eventually in deciding on an approach to therapy in the clinic.

Conclusions The treatment of patients with SLE should be multi- layered and take into account the prevention of disease progression, relapse, damage and comorbidities from the time of diagnosis. Indeed, the more timely the treat- ment, the greater the likelihood of achieving control of disease with a governable drug regimen in a reasonable time span.

From the clinical point of view, validating the optimal dosages and administration protocols for the available drugs, including glucocorticoids, through RCTs would be opportune, as the drawbacks of overtreatment are widely acknowledged. Accordingly, trials that investigate the usefulness of modifying and/or tapering therapy fol- lowing the achievement of treatment targets are needed, and the resulting strategies must be agreed upon before the concept of treating patients to a target can be imple- mented in clinical practice, without which patients might be subjected to unnecessary long- standing exposure to toxic drugs. In this view, patient- tailored outcomes should be used in clinical trials whenever possible.

Alongside the need to agree upon a shared definition of treatment targets, including remission and LDA, it is important to precisely define the concept of refractoriness to standard- of-care therapy, which is often required for access to biologic therapies. The judgement of the physi- cian and economic considerations will always be involved in treatment decisions; nevertheless, we should realize that patients with mild to moderate SLE who do not respond to treatment with glucocorticoids and antimalarial drugs might already be considered refractory to standard- of-care therapy and as such might be elected for biologic therapy to benefit from early treatment. Additionally, the view of preventing disease- related or treatment- related flares and damage in the long term should always be factored into the choice of therapeutic strategy.

Although personalized medicine on the basis of molecular stratification is not yet widespread, treatments should be tailored to the patient to increase the likeli- hood of controlling disease and promote patient adher- ence. In this regard, studies aimed at evaluating the real effect of patient stratification on treatment efficacy are truly necessary.

Published online 11 December 2018

1. Doria, A. et al. Long- term prognosis and causes of death in systemic lupus erythematosus. Am. J. Med. 119, 700–706 (2006).

2. Jorge, A. M., Lu, N., Zhang, Y., Rai, S. K. & Choi, H. K. Unchanging premature mortality trends in systemic lupus erythematosus: a general population- based study (1999–2014). Rheumatology (Oxford) 57, 337–344 (2018).

3. Yen, E. Y. & Singh, R. R. Lupus- an unrecognized leading cause of death in young females: a population- based study using nationwide death certificates,

2000–2015. Arthritis. Rheumatol. 70, 1251–1255 (2018).

4. Van Vollenhoven, R. F. et al. Treat- to-target in systemic lupus erythematosus: recommendations from an international task force. Ann. Rheum. Dis. 73, 958–967 (2014).

5. Little, J. et al. Glucocorticoid use and factors associated with variability in this use in the Systemic Lupus International Collaborating Clinics Inception Cohort. Rheumatology (Oxford) 57, 677–687 (2018).

6. Wallace, D. J. et al. Systemic lupus erythematosus- survival patterns. Experience with 609 patients. JAMA 245, 934–938 (1981).

7. Doria, A. et al. SLE diagnosis and treatment: when early is early. Autoimmun. Rev. 10, 55–60 (2010).

8. Nightingale, A. L., Davidson, J. E., Molta, C. T., Kan, H. J. & McHugh, N. J. Presentation of SLE in UK primary care using the Clinical Practice Research Datalink. Lupus. Sci. Med. 4, e000172 (2017).

9. Oglesby, A. et al. Impact of early versus late systemic lupus erythematosus diagnosis on clinical and

w w w.nature.com/nrrheum

R e v i e w s

44 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

economic outcomes. Appl. Health. Econ. Health. Policy. 12, 179–190 (2014).

10. Arbuckle, M. R. et al. Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N. Engl. J. Med. 349, 1526–1533 (2003).

11. Rees, F. et al. Early clinical features in systemic lupus erythematosus: can they be used to achieve earlier diagnosis? A risk prediction model. Arthritis Care Res. (Hoboken) 69, 833–841 (2017).

12. Arriens, C., Wren, J. D., Munroe, M. E. & Mohan, C. Systemic lupus erythematosus biomarkers: the challenging quest. Rheumatology (Oxford) 56, i32–i45 (2017).

13. Compagno, M. et al. Low diagnostic and predictive value of anti- dsDNA antibodies in unselected patients with recent onset of rheumatic symptoms: results from a long- term follow- up Scandinavian multicentre study. Scand. J. Rheumatol. 42, 311–316 (2013).

14. Piga, M. et al. Failure to achieve lupus low disease activity state (LLDAS) six months after diagnosis is associated with early damage accrual in Caucasian patients with systemic lupus erythematosus. Arthritis. Res. Ther. 19, 247 (2017).

15. Nossent, J. et al. Disease activity and damage accrual during the early disease course in a multinational inception cohort of patients with systemic lupus erythematosus. Lupus 19, 949–956 (2010).

16. Iaccarino, L. et al. Clinical predictors of response and discontinuation of belimumab in patients with systemic lupus erythematosus in real life setting. Results of a large, multicentric, nationwide study. J. Autoimmun. 86, 1–8 (2018). This paper presents the first large, prospective, multicentre study on the real- life use of belimumab in SLE and its clinical effects.

17. Esdaile, J. M., Joseph, L., MacKenzie, T., Kashgarian, M. & Hayslett, J. P. The benefit of early treatment with immunosuppressive agents in lupus nephritis. J. Rheumatol. 21, 2046–2051 (1994).

18. Jacobsen, S. et al. Prognostic value of renal biopsy and clinical variables in patients with lupus nephritis and normal serum creatinine. Scand. J. Rheumatol. 28, 288–299 (1999).

19. Faurschou, M., Starklint, H., Halberg, P. & Jacobsen, S. Prognostic factors in lupus nephritis: diagnostic and therapeutic delay increases the risk of terminal renal failure. J. Rheumatol. 33, 1563–1569 (2006).

20. Ciruelo, E., De la Cruz, J., López, I. & Gómez- Reino, J. J. Cumulative rate of relapse of lupus nephritis after successful treatment with cyclophosphamide. Arthritis. Rheum. 39, 2028–2034 (1996).

21. Fiehn, C. et al. Improved clinical outcome of lupus nephritis during the past decade: importance of early diagnosis and treatment. Ann. Rheum. Dis. 62, 435–439 (2003).

22. Munroe, M. E. et al. Altered type II interferon precedes autoantibody accrual and elevated type I interferon activity prior to systemic lupus erythematosus classification. Ann. Rheum. Dis. 75, 2014–2021 (2016). This paper highlights the occurrence of immunological abnormalities in the very early stages of SLE development and sheds new light on pathogenesis.

23. Lu, R. et al. Dysregulation of innate and adaptive serum mediators precedes systemic lupus erythematosus classification and improves prognostic accuracy of autoantibodies. J. Autoimmun. 74, 182–193 (2016).

24. Calixto, O. J., Franco, J. S. & Anaya, J. M. Lupus mimickers. Autoimmun. Rev. 13, 865–872 (2014).

25. Doria, A. & Briani, C. Primary prevention of systemic lupus erythematosus. Nat. Clin. Pract. Rheumatol. 4, 576–577 (2008).

26. Doria, A. & Briani, C. Lupus: improving long- term prognosis. Lupus 17, (166–170 (2008).

27. Munroe, M. E. et al. Discerning risk of disease transition in relatives of systemic lupus erythematosus patients utilizing soluble mediators and clinical features. Arthritis Rheumatol. 69, 630–642 (2017). This paper explores the topic of anticipating the onset of systemic autoimmunity in SLE, which might shift the border of prevention and treatment.

28. Durcan, L. & Petri, M. Immunomodulators in SLE: clinical evidence and immunologic actions. J. Autoimmun. 74, 73–84 (2016).

29. Bizzarro, N. et al. Anti- prothrombin antibodies predict thrombosis in patients with systemic lupus erythematosus: a 15-year longitudinal study. J. Thromb. Haemost. 5, 1158–1164 (2007).

30. Pengo, V. et al. Efficacy and safety of rivaroxaban versus warfarin in high- risk patients with antiphospholipid syndrome: rationale and design of the Trial on Rivaroxaban in AntiPhospholipid Syndrome (TRAPS) trial. Lupus 25, 301–306 (2016).

31. Arnaud, L. et al. Patient- level analysis of five international cohorts further confirms the efficacy of aspirin for the primary prevention of thrombosis in patients with antiphospholipid antibodies. Autoimmun. Rev. 14, 192–200 (2015).

32. Ruiz- Irastorza, G. et al. Evidence- based recommendations for the prevention and long- term management of thrombosis in antiphospholipid antibody- positive patients: report of a task force at the 13th international congress on antiphospholipid antibodies. Lupus 20, 206–218 (2011).

33. Erkan, D. et al. Aspirin for primary thrombosis prevention in the antiphospholipid syndrome: a randomized, double- blind, placebo- controlled trial in asymptomatic antiphospholipid antibody- positive individuals. Arthritis Rheum. 56, 2382–2391 (2007).

34. Moroni, G. et al. The long- term outcome of 93 patients with proliferative lupus nephritis. Nephrol. Dial. Transplant. 22, 2531–2539 (2007).

35. Lateef, A. & Petri, M. Unmet medical needs in systemic lupus erythematosus. Arthritis. Res. Ther. 14 (Suppl. 4), S4 (2012).

36. Doria, A. et al. Annual direct medical cost of active systemic lupus erythematosus in five European countries. Ann. Rheum. Dis. 73, 154–160 (2014).

37. Moroni, G. et al. Changing patterns in clinical- histological presentation and renal outcome over the last five decades in a cohort of 499 patients with lupus nephritis. Ann. Rheum. Dis. 77, 1318–1325 (2018). This study explores the changes in patterns of demographic, clinical and histological presentation in lupus nephritis over five decades, the longest observational period of its kind.

38. Golder, V. et al. Frequency and predictors of the lupus low disease activity state in a multi- national and multi- ethnic cohort. Arthritis. Res. Ther. 18, 260 (2016).

39. van Vollenhoven, R. et al. A framework for remission in SLE: consensus findings from a large international task force on definitions of remission in SLE (DORIS). Ann. Rheum. Dis. 76, 554–561 (2017).

40. Zen, M. et al. Prolonged remission in Caucasian patients with SLE: prevalence and outcomes. Ann. Rheum. Dis. 74, 2117–2122 (2015).

41. Polachek, A., Gladman, D. D., Su, J. & Urowitz, M. B. Defining low disease activity in systemic lupus erythematosus. Arthritis Care Res. (Hoboken) 69, 997–1003 (2017).

42. Franklyn, K. et al. Definition and initial validation of a Lupus Low Disease Activity State (LLDAS). Ann. Rheum. Dis. 75, 1615–1621 (2016). This paper is the first to propose a definition of LDA in SLE.

43. Ugarte- Gil, M. F. et al. Remission and Low Disease Activity Status (LDAS) protect lupus patients from damage occurrence: data from a multiethnic, multinational Latin American lupus cohort (GLADEL). Ann. Rheum. Dis. 76, 2071–2074 (2017).

44. Zen, M. et al. The effect of different durations of remission on damage accrual: results from a prospective monocentric cohort of Caucasian patients. Ann. Rheum. Dis. 75, 562–565 (2017). This paper is the first to identify a clear period of remission that is associated with protection from organ damage.

45. Zen, M. et al. Lupus low disease activity state is associated with a decrease in damage progression in Caucasian patients with SLE, but overlaps with remission. Ann. Rheum. Dis. 77, 104–110 (2018).

46. Mok, C. C. et al. Prevalence of remission and its effect on damage and quality of life in Chinese patients with systemic lupus erythematosus. Ann. Rheum. Dis. 76, 1420–1425 (2017).

47. Tsang- A-Sjoe, M. W. et al. Both prolonged remission and lupus low disease activity state are associated with reduced damage accrual in systemic lupus erythematosus. Rheumatology (Oxford) 56, 121–128 (2017).

48. Petri, M. & Magder, L. S. Comparison of remission and lupus low disease activity state in damage prevention in a United States systemic lupus erythematosus cohort. Arthritis Rheumatol. 70, 1790–1795 (2018).

49. Formiga, F. et al. High disease activity at baseline does not prevent a remission in patients with systemic lupus erythematosus. Rheumatology (Oxford) 38, 724–727 (1999).

50. Medina- Quiñones, C. V., Ramos- Merino, L., Ruiz- Sada, P. & Isenberg, D. Analysis of complete remission in systemic lupus erythematosus patients over a 32-year period. Arthritis Care Res. (Hoboken) 68, 981–987 (2016).

51. Urowitz, M. B., Feletar, M., Bruce, I. N., Ibañez, D. & Gladman, D. D. Prolonged remission in systemic lupus erythematosus. J. Rheumatol. 32, 1467–1472 (2005).

52. Drenkard, C., Villa, A. R., Garcia- Padilla, C., Pérez-Vázquez, M. E. & Alarcón- Segovia, D. Remission of systematic lupus erythematosus. Medicine (Baltimore) 75, 88–98 (1996).

53. Isenberg, D. A. et al. An assessment of disease flare in patients with systemic lupus erythematosus: a comparison of BILAG 2004 and the flare version of SELENA. Ann. Rheum. Dis. 70, 54–59 (2011).

54. Aranow, C. A pilot study to determine the optimal timing of the Physician Global Assessment (PGA) in patients with systemic lupus erythematosus. Immunol. Res. 63, 167–169 (2015).

55. Bertsias, G. et al. Joint European League Against Rheumatism and European Renal Association- European Dialysis and Transplant Association (EULAR/ ERA- EDTA) recommendations for the management of adult and paediatric lupus nephritis. Ann. Rheum. Dis. 71, 1771–1782 (2012).

56. Chen, Y. E., Korbet, S. M., Katz, R. S., Schwartz, M. M. & Lewis, E. J. Value of a complete or partial remission in severe lupus nephritis. Clin. J. Am. Soc. Nephrol. 3, 46–53 (2008).

57. Houssiau, F. A. et al. Early response to immunosuppressive therapy predicts good renal outcome in lupus nephritis: lessons from long- term followup of patients in the Euro- Lupus Nephritis Trial. Arthritis. Rheum. 50, 3934–3940 (2004).

58. Medina- Rosas, J., Yap, K. S., Anderson, M., Su, J. & Touma, Z. Utility of urinary protein- creatinine ratio and protein content in a 24-hour urine collection in systemic lupus erythematosus: a systematic review and meta- analysis. Arthritis Care Res. (Hoboken) 68, 1310–1319 (2016).

59. Houssiau, F. A. et al. Immunosuppressive therapy in lupus nephritis: the Euro- Lupus Nephritis Trial, a randomized trial of low- dose versus high- dose intravenous cyclophosphamide. Arthritis Rheum. 46, 2121–2131 (2002).

60. Dall’Era, M. et al. Predictors of long- term renal outcome in lupus nephritis trials: lessons learned from the Euro- Lupus Nephritis cohort. Arthritis Rheumatol. 67, 1305–1313 (2015).

61. Tamirou, F. et al. MAINTAIN nephritis trial investigators. A proteinuria cut- off level of 0.7 g/day after 12 months of treatment best predicts long- term renal outcome in lupus nephritis: data from the MAINTAIN nephritis trial. Lupus. Sci. Med. 12, e000123 (2015).

62. Arriens, C. et al. Prognostic significance of repeat biopsy in lupus nephritis: histopathologic worsening and a short time between biopsies is associated with significantly increased risk for end stage renal disease and death. Clin. Immunol. 185, 3–9 (2017).

63. Moroni, G., Depetri, F. & Ponticelli, C. Lupus nephritis: when and how often to biopsy and what does it mean? J. Autoimmun. 74, 27–40 (2016).

64. Touma, Z. et al. Time to recovery from proteinuria in patients with lupus nephritis receiving standard treatment. J. Rheumatol. 41, 688–697 (2014).

65. Moroni, G. et al. Membranous nephropathy in systemic lupus erythematosus: long- term outcome and prognostic factors of 103 patients. Semin. Arthritis Rheum. 41, 642–651 (2012).

66. Moroni, G. et al. Withdrawal of therapy in patients with proliferative lupus nephritis: long- term follow- up. Nephrol. Dial. Transplant. 21, 1541–1548 (2006).

67. Mosca, M., Tani, C. & Aringer, M. Withdrawal of therapy in non- renal systemic lupus erythematosus: is this an achievable goal? Clin. Exp. Rheumatol. 31 (Suppl. 78), S71–S74 (2013).

68. Gordon, C. et al. British Society for Rheumatology standards, Audit and Guidelines Working Group. The British Society for Rheumatology guideline for the management of systemic lupus erythematosus in adults. Rheumatolohy (Oxford) 57, e1–e45 (2018).

69. Bertsias, G. K. et al. EULAR recommendations for the management of systemic lupus erythematosus with neuropsychiatric manifestations: report of a task force of the EULAR standing committee for clinical affairs. Ann. Rheum. Dis. 69, 2074–2082 (2010).

70. Pons- Estel, B. A. et al. Grupo Latino Americano de Estudio del Lupus (GLADEL) and Pan- American League of Associations of Rheumatology (PANLAR).

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 45

First Latin American clinical practice guidelines for the treatment of systemic lupus erythematosus: Latin American Group for the Study of Lupus (GLADEL, Grupo Latino Americano de Estudio del Lupus)-Pan- American League of Associations of Rheumatology (PANLAR). Ann. Rheum. Dis. 77, 1549–1557 (2018).

71. Posnick, J. Systemic lupus erythematosus. The effect of corticotropin and adrenocorticoid therapy on survival rate. Calif. Med. 98, 308–312 (1963).

72. Merrill, J. T. et al. The biomarkers of lupus disease study: a bold approach may mitigate interference of background immunosuppressants in clinical trials. Arthritis Rheumatol. 69, 1257–1266 (2017). This study explores the consequences of immunosuppression- free regimens in SLE and indirectly shows the insufficient effects of corticosteroids in maintaining long- term treatment responses.

73. Bruce, I. N. et al. Factors associated with damage accrual in patients with systemic lupus erythematosus: results from the Systemic Lupus International Collaborating Clinics (SLICC) Inception Cohort. Ann. Rheum. Dis. 74, 1706–1713 (2015).

74. Al Sawah, S. et al. Effect of corticosteroid use by dose on the risk of developing organ damage over time in systemic lupus erythematosus- the Hopkins Lupus Cohort. Lupus. Sci. Med. 2, e000066 (2015).

75. Guiducci, C. et al. TLR recognition of self nucleic acids hampers glucocorticoid activity in lupus. Nature 465, 937–941 (2010).

76. Edwards, J. C., Snaith, M. L. & Isenberg, D. A. A double blind controlled trial of methylprednisolone infusions in systemic lupus erythematosus using individualized outcome assessment. Ann. Rheum. Dis. 46, 773–776 (1987).

77. Fischer- Betz, R. et al. Renal outcome in patients with lupus nephritis using a steroid- free regimen of monthly intravenous cyclophosphamide: a prospective observational study. J. Rheumatol. 39, 2111–2117 (2012).

78. Ruiz- Irastorza, G. et al. Repeated pulses of methyl- prednisolone with reduced doses of prednisone improve the outcome of class III, IV and V lupus nephritis: an observational comparative study of the Lupus- Cruces and lupus- Bordeaux cohorts. Autoimmun. Rev. 16, 826–832 (2017).

79. Zeher, M. et al. Efficacy and safety of enteric- coated mycophenolate sodium in combination with two glucocorticoid regimens for the treatment of active lupus nephritis. Lupus 20, 1484–1493 (2011).

80. Merrill, J. T. et al. Lupus community panel proposals for optimising clinical trials: 2018. Lupus. Sci. Med. 5, e000258 (2018).

81. Lightstone, L. et al. Can we manage lupus nephritis without chronic corticosteroids administration? Autoimmun. Rev. 17, 4–10 (2018). This commentary provides a complete overview of the evidence supporting or cautioning against steroid- free regimens in lupus nephritis.

82. Condon, M. B. et al. Prospective observational single- centre cohort study to evaluate the effectiveness of treating lupus nephritis with rituximab and mycophenolate mofetil but no oral steroids. Ann. Rheum. Dis. 72, 1280–1286 (2013).

83. Porter, A. B. et al. Long term follow up of the Rituxilup steroid sparing regimen in lupus nephritis [abstract FR- OR065]. J. Am. Soc. Nephrol. 25, 61A (2014).

84. Pepper, R. et al. Rituximab is an effective treatment for lupus nephritis and allows a reduction in maintenance steroids. Nephrol. Dial. Transplant. 24, 3717–3723 (2009).

85. Pillay, C., Levy, J. B., Cairns, T. & Lightstone, L. G. M. Treating lupus nephritis with rituximab and mycophenolate mofetil (rituxirescue regimen) without increasing maintenance oral steroids leads to sustained disease remission and steroid reduction. J. Am. Soc. Nephrol. 27, 500A (2016).

86. Iaccarino, L. et al. Effects of belimumab on flare rate and expected damage progression in patients with active systemic lupus erythematosus. Arthritis Care Res. (Hoboken) 69, 115–123 (2017).

87. Ginzler, E. M. et al. Disease control and safety of belimumab plus standard therapy over 7 years in patients with systemic lupus erythematosus. J. Rheumatol. 41, 300–309 (2014).

88. Trentin, F. et al. Effectiveness, tolerability, and safety of belimumab in patients with refractory SLE: a review of observational clinical- practice-based studies. Clin. Rev. Allergy Immunol. 54, 331–343 (2018).

89. Bruce, I. N. et al. Long- term organ damage accrual and safety in patients with SLE treated with belimumab plus standard of care. Lupus 25, 699–709 (2016).

90. Liu, Z. et al. Multitarget therapy for induction treatment of lupus nephritis: a randomized trial. Ann. Intern. Med. 162, 18–26 (2015).

91. Zhang, H. et al. Multitarget therapy for maintenance treatment of lupus nephritis. J. Am. Soc. Nephrol. 28, 3671–3678 (2017).

92. Furie, R. et al. Anifrolumab, an anti- interferon-α receptor monoclonal antibody, in moderate- to-severe systemic lupus erythematosus. Arthritis Rheumatol. 69, 376–386 (2017).

93. US National Library of Medicine. ClinicalTrials.gov https://clinicaltrials.gov/ct2/show/ NCT03021499?term=NCT03021499&rank=1 (2018).

94. Watanabe, H. et al. The efficacy of add- on tacrolimus for minor flare in patients with systemic lupus erythematosus: a retrospective study. Lupus 25, 54–60 (2016).

95. Mok, C. C. Calcineurin inhibitors in systemic lupus erythematosus. Best Pract. Res. Clin. Rheumatol. 31, 429–438 (2017).

96. Gracia- Tello, B. et al. The use of rituximab in newly diagnosed patients with systemic lupus erythematosus: long- term steroid saving capacity and clinical effectiveness. Lupus. Sci. Med. 4, e000182 (2017).

97. Roccatello, D. et al. Intensive short term treatment with rituximab, cyclophosphamide and methylprednisolone pulses induces remission in severe cases of SLE with nephritis and avoids further immunosuppressive maintenance therapy. Nephrol. Dial. Transplant. 26, 3987–3989 (2011).

98. Ruiz- Arruza, I. et al. Comparison of high versus low-medium prednisone doses for the treatment of systemic lupus erythematosus patients with high disease activity at diagnosis. Autoimmun. Rev. 14, 875–879 (2015). This study shows that low- dose steroids are as effective as high- dose steroids during induction treatment and have a better safety profile.

99. Gladman, D. D., Iban˜ez, D., Ruiz, I. & Urowitz, M. B. Recommendations for frequency of visits to monitor systemic lupus erythematosus in asymptomatic patients: data from an observational cohort study. J. Rheumatol. 40, 630–633 (2013).

100. Bertsias, G. et al. Task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics. EULAR recommendations for the management of systemic lupus erythematosus. Report of a Task Force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics. Ann. Rheum. Dis. 67, 195–205 (2008).

101. Ruiz- Irastorza, G., Ramos- Casals, M., Brito- Zeron, P. & Khamashta, M. A. Clinical efficacy and side effects of antimalarials in systemic lupus erythematosus: a systematic review. Ann. Rheum. Dis. 69, 20–28 (2010).

102. Hsu, C. Y. et al. Adherence to hydroxychloroquine improves long- term survival of patients with systemic lupus erythematosus. Rheumatology (Oxford) 57, 1743–1751 (2018).

103. Alarcón, G. S. et al. Effect of hydroxychloroquine on the survival of patients with systemic lupus erythematosus: data from LUMINA, a multiethnic US cohort (LUMINA L). Ann. Rheum. Dis. 66, 1168–1172 (2007).

104. Costedoat- Chalumeau, N. et al. Low blood concentration of hydroxychloroquine is a marker for and predictor of disease exacerbations in patients with systemic lupus erythematosus. Arthritis. Rheum. 54, 3284–3290 (2006).

105. Costedoat- Chalumeau, N. et al. Routine hydroxychloroquine blood concentration measurement in systemic lupus erythematosus reaches adulthood. J. Rheumatol. 42, 1997–1999 (2015).

106. Ting, T. V. et al. Usefulness of cellular text messaging for improving adherence among adolescents and young adults with systemic lupus erythematosus. J. Rheumatol. 39, 174–179 (2012).

107. Costedoat- Chalumeau, N. et al. A prospective international study on adherence to treatment in 305 patients with flaring SLE: assessment by drug levels and self- administered questionnaires. Clin. Pharmacol. Ther. 103, 1074–1082 (2018).

108. Durcan, L., Clarke, W. A., Magder, L. S. & Petri, M. Hydroxychloroquine blood levels in systemic lupus erythematosus: clarifying dosing controversies and improving adherence. J. Rheumatol. 42, 2092–2097 (2015).

109. Fessler, B. J. et al. Systemic lupus erythematosus in three ethnic groups: XVI. Association of hydroxychloroquine use with reduced risk of damage accrual. Arthritis Rheum. 52, 1473–1480 (2005).

110. Pons- Estel, G. et al. Protective effect of hydroxychloroquine on renal damage in patients with lupus nephritis: LXV, data from a multiethnic US cohort. Arthritis Rheum. 61, 830–839 (2009).

111. Londoño Jimenez, A. et al. Tubulointerstitial damage in lupus nephritis: a comparison of the factors associated with tubulointerstitial inflammation and renal scarring. Arthritis Rheumatol. 70, 1801–1806 (2018).

112. Pego- Reigosa, J. M. et al. Efficacy and safety of nonbiologic immunosuppressants in the treatment of nonrenal systemic lupus erythematosus: a systematic review. Arthritis Care Res. (Hoboken) 65, 1775–1785 (2013).

113. Allison, A. C. & Eugui, E. M. Purine metabolism and immunosuppressive effects of mycophenolate mofetil (MMF). Clin. Transplant. 10, 77–84 (1996).

114. Chen, W. et al. Outcomes of maintenance therapy with tacrolimus versus azathioprine for active lupus nephritis: a multicenter randomized clinical trial. Lupus 21, 944–952 (2012).

115. Dooley, M. A. et al. Speed of remission with the use of voclosporin, MMF and low dose steroids: results of a global lupus nephritis study [abstract]. Arthritis Rheumatol. 68 (Suppl. 10), 5L (2016).

116. Fu, J. et al. Transcriptomic analysis uncovers novel synergistic mechanisms in combination therapy for lupus nephritis. Kidney. Int. 93, 416–429 (2018). The researchers of this study, in a transcriptomic analysis, shed light on the synergistic effects of combination therapy.

117. Kraaij, T. et al. TAC- TIC use of tacrolimus- based regimens in lupus nephritis. Lupus Sci. Med. 3, e000169 (2016).

118. van Vollenhoven, R. et al. Efficacy and safety of ustekinumab, an interleukin 12/23 inhibitor, in patients with active systemic lupus erythematosus: results of a phase 2, randomized placebo- controlled study [abstract]. Arthritis. Rheumatol. 69 (Suppl. 10), 6L (2017).

119. Wallace, D. J. et al. Baricitinib for systemic lupus erythematosus: a double- blind, randomised, placebo- controlled, phase 2 trial. Lancet 392, 222–231 (2018).

120. Furie, R. et al. A phase III, randomized, placebo- controlled study of belimumab, a monoclonal antibody that inhibits B lymphocyte stimulator, in patients with systemic lupus erythematosus. Arthritis Rheum. 63, 3918–3930 (2011).

121. Navarra, S. V. et al. Efficacy and safety of belimumab in patients with active systemic lupus erythematosus: a randomised, placebo- controlled, phase 3 trial. Lancet 377, 721–731 (2011).

122. Merrill, J. T. et al. Efficacy and safety of rituximab in moderately- to-severely active systemic lupus erythematosus: the randomized, double- blind, phase II/III systemic lupus erythematosus evaluation of rituximab trial. Arthritis Rheum. 62, 222–233 (2010).

123. Rovin, B. H. et al. Efficacy and safety of rituximab in patients with active proliferative lupus nephritis: the lupus nephritis assessment with rituximab study. Arthritis Rheum. 64, 1215–1226 (2012).

124. Iaccarino, L. et al. Efficacy and safety of off- label use of rituximab in refractory lupus: data from the Italian Multicentre Registry. Clin. Exp. Rheumatol. 33, 449–456 (2015).

125. McCarthy, E. M. et al. Short- term efficacy and safety of rituximab therapy in refractory systemic lupus erythematosus: results from the British Isles lupus assessment group biologics register. Rheumatology (Oxford) 57, 470–479 (2018).

126. Witt, M. et al. Clinical outcomes and safety of rituximab treatment for patients with systemic lupus erythematosus (SLE) - results from a nationwide cohort in Germany (GRAID). Lupus 22, 1142–1149 (2013).

127. Hahn, B. et al. American College of Rheumatology guidelines for screening, treatment, and management of lupus nephritis. Arthritis Care Res. (Hoboken) 64, 797–808 (2012).

128. Jovancevic, B., Lindholm, C. & Pullerits, R. Anti B cell therapy against refractory thrombocytopenia in SLE and MCTD patients: long- term follow- up and review of the literature. Lupus 22, 664–674 (2013).

129. Hofmann, S. C., Leandro, M. J., Morris, S. D. & Isenberg, D. A. Effects of rituximab- based B cell depletion therapy on skin manifestations of lupus erythematosus—report of 17 cases and review of the literature. Lupus 22, 932–939 (2013).

130. Weidenbusch, M., Römmele, C., Schröttle, A. & Anders, H. J. Beyond the LUNAR trial. Efficacy of rituximab in refractory lupus nephritis. Nephrol. Dial. Transplant. 28, 106–111 (2013).

w w w.nature.com/nrrheum

R e v i e w s

46 | J A N UA RY 2 0 1 9 | v o l U m e 1 5

131. Ezeonyeji, A. N. & Isenberg, D. A. Early treatment with rituximab in newly diagnosed systemic lupus erythematosus patients: a steroid- sparing regimen. Rheumatology (Oxford) 51, 476–481 (2012).

132. Stohl, W. et al. Efficacy and safety of subcutaneous belimumab in systemic lupus erythematosus: a fifty- two-week randomized, double- blind, placebo- controlled study. Arthritis Rheumatol. 69, 1016–1027 (2017).

133. Fanouriakis, A. et al. Low disease activity- irrespective of serologic status at baseline- associated with reduction of corticosteroid dose and number of flares in patients with systemic lupus erythematosus treated with belimumab: a real- life observational study. Semin. Arthritis Rheum. https://doi.org/10.1016/j. semarthrit.2018.02.014 (2018).

134. US National Library of Medicine. ClinicalTrials.gov https://clinicaltrials.gov/ct2/show/ NCT01639339?term=NCT01639339&rank=1 (2018).

135. Dooley, M. A. et al. Effect of belimumab treatment on renal outcomes: results from the phase 3 belimumab clinical trials in patients with SLE. Lupus 22, 63–72 (2013).

136. Kraaij, T., Huizinga, T. W., Rabelink, T. J. & Teng, Y. K. Belimumab after rituximab as maintenance therapy in lupus nephritis. Rheumatology (Oxford) 53, 2122–2124 (2014).

137. Simonetta, F., Allali, D., Roux- Lombard, P. & Chizzolini, C. Successful treatment of refractory lupus nephritis by the sequential use of rituximab and belimumab. Joint Bone Spine 84, 235–236 (2017).

138. Psarelis, S., Nikiphorou, E. & Boumpas, D. T. Successful use of sequential B cell depletion therapy in lupus. Lupus 27, 345–346 (2018).

139. Carter, L. M., Isenberg, D. A. & Ehrenstein, M. R. Elevated serum BAFF levels are associated with rising anti- double-stranded DNA antibody levels and disease flare following B cell depletion therapy in systemic lupus erythematosus. Arthritis Rheum. 65, 2672–2679 (2013).

140. US National Library of Medicine. ClinicalTrials.gov https://clinicaltrials.gov/ct2/show/NCT03312907? term=NCT03312907&rank=1 (2018).

141. Specchia, M. L. et al. Health technology assessment of belimumab: a new monoclonal antibody for the treatment of systemic lupus erythematosus. Biomed. Res. Int. 2014, 704207 (2014).

142. Vallejo- Aparicio, L. A., Díaz- Cerezo, S., Parrondo, J. & García- Aparicio, A. M. Cost- effectiveness analysis of belimumab in the treatment of adult systemic lupus erythematosus (Sle) Patients With Positive Biomarkers In Spain. Value. Health. 17, A530 (2014).

143. Athanasakis, K. et al. Belimumab for the treatment of systemic lupus erythematosus (Sle) in Greece: a cost- effectiveness and cost- utility analysis. Value Health 17, A532–A533 (2014).

144. National Institute for Health and Care Excellence. Final appraisal determination: belimumab for treating active autoantibody- positive systemic lupus erythematosus. NICE https://www.nice.org.uk/guidance/ta397/ documents/final- appraisal-determination- document (2016).

145. European Medicines Agency. Benlysta: EPAR- product information. Annex I: summary of product characteristics. ema.europa https://www.ema.europa.eu/ documents/product- information/benlysta- epar-product- information_en.pdf (2017).

146. van Vollenhoven, R. F. et al. Belimumab in the treatment of systemic lupus erythematosus: high disease activity predictors of response. Ann. Rheum. Dis. 71, 1343–1349 (2012).

147. Dooley, M. A. et al. Mycophenolate versus azathioprine as maintenance therapy for lupus nephritis. N. Engl. J. Med. 365, 1886–1895 (2011).

148. Houssiau, F. A. et al. Azathioprine versus mycophenolate mofetil for long- term immunosuppression in lupus nephritis: results from the MAINTAIN Nephritis Trial. Ann. Rheum. Dis. 69, 2083–2089 (2010).

149. Zen, M. et al. Disease activity patterns in a monocentric cohort of SLE patients: a seven- year follow- up study. Clin. Exp. Rheumatol. 30, 856–863 (2012).

150. Györi, N. et al. Disease activity patterns over time in patients with SLE: analysis of the Hopkins Lupus Cohort. Lupus. Sci. Med. 4, e000192 (2017).

151. Bakker, M. F., Jacobs, J. W., Verstappen, S. M. & Bijlsma, J. W. Tight control in the treatment of rheumatoid arthritis: efficacy and feasibility. Ann. Rheum. Dis. 66 (Suppl. 3), 56–60 (2007).

152. Aljohani, R., Gladman, D. D., Su, J. & Urowitz, M. B. Comparison of systemic lupus erythematosus (SLE) patients managed early after diagnosis in specialty versus community care clinics. Clin. Rheumatol. 36, 1773–1778 (2017).

153. Steiman, A. J. et al. Outcomes in patients with systemic lupus erythematosus with and without a prolonged serologically active clinically quiescent period. Arthritis Care Res. 64, 511–518 (2012).

154. Zahr, Z. A., Fang, H., Magder, L. S. & Petri, M. Predictors of corticosteroid tapering in SLE patients: the Hopkins Lupus Cohort. Lupus 22, 697–701 (2013).

155. Mosca, M., Tani, C., Carli, L. & Bombardieri, S. Glucocorticoids in systemic lupus erythematosus. Clin. Exp. Rheumatol. 29 (Suppl. 68), S126–S129 (2011).

156. The Canadian Hydroxychloroquine Study Group. A randomized study of the effect of withdrawing hydroxychloroquine sulfate in systemic lupus erythematosus. N. Engl. J. Med. 324, 150–482 (1991).

157. Levy, R. et al. Hydroxychloroquine (HCQ) in lupus pregnancy: double- blind and placebo controlled study. Lupus 10, 401–404 (2001).

158. Tsakonas, E. et al. A long- term study of hydroxychloroquine withdrawal on exacerbations in systemic lupus erythematosus. The Canadian Hydroxychloroquine Study Group. Lupus. 7, 80–85 (1998).

159. Costedoat- Chalumeau, N. et al. Hydroxychloroquine in systemic lupus erythematosus: results of a French multicentre controlled trial (PLUS Study). Ann. Rheum. Dis. 72, 1786–1792 (2013).

160. Mok, C. C. et al. Hydroxychloroquine serum concentrations and flares of systemic lupus erythematosus: a longitudinal cohort analysis. Arthritis Care Res. (Hoboken) 68, 1295–1302 (2016).

161. Cunha, C. et al. Hydroxycloroquine blood concentration in lupus nephritis: a determinant of disease outcome? Nephrol. Dial. Transplant. 33, 1604–1610 (2017).

162. Ugarte, A., Danza, A. & Ruiz- Irastorza, G. Glucocorticoids and antimalarials in systemic lupus erythematosus: an update and future directions. Curr. Opin. Rheumatol. 30, 482–489 (2018).

163. Sharon, E., Kaplan, D. & Diamond, H. S. Exacerbation of systemic lupus erythematosus after withdrawal of azathioprine therapy. N. Engl. J. Med. 288, 122–124 (1973).

164. Nikpour, M., Urowitz, M., Ibanez, D. & Gladman, D. D. Frequency and determinants of flare and persistently active disease in systemic lupus erythematosus. Arthritis Rheum. 61, 1152–1158 (2009).

165. Pablos, J. L., Gutierrez- Millet, V. & Gomez- Reino, J. J. Remission of lupus nephritis with cyclophosphamide and late relapses following therapy withdrawal. Scand. J. Rheumatol. 23, 142–144 (1994).

166. Mosca, M. et al. Therapy with pulse methylprednisolone and short course pulse cyclophosphamide for diffuse proliferative glomerulonephritis. Lupus 10, 253–225 (2001).

167. Moroni, G., Longhi, S., Gliglio, E., Messa, P. & Ponticelli, C. What happens after complete withdrawal of therapy in patients with lupus nephritis. Clin. Exp. Rheumatol. 31, S75–S81 (2013).

168. Barturen, G. & Alarcón- Riquelme, M. E. Systemic lupus erythematosus in 2016: gene expression profiling comes closer to the clinic. Nat. Rev. Rheumatol. 13, 69–70 (2017).

169. Pirone, C. et al. Predictive and prognostic factors influencing outcomes of rituximab therapy in systemic lupus erythematosus (SLE): a systematic review. Semin. Arthritis Rheum. 47, 384–396 (2017).

170. Dall’Era, M., Stone, D., Levesque, V., Cisternas, M. & Wofsy, D. Identification of biomarkers that predict response to treatment of lupus nephritis with mycophenolate mofetil or pulse cyclophosphamide. Arthritis Care Res. (Hoboken) 63, 351–357 (2011).

171. Gottenberg, J. E. et al. Efficacy of epratuzumab, an anti- CD22 monoclonal IgG antibody, in systemic lupus erythematosus patients with associated sjögren’s syndrome: post Hoc analyses from the EMBODY trials. Arthritis Rheumatol. 70, 763–773 (2018).

172. Teruel, M., Chamberlain, C. & Alarcón- Riquelme, M. E. Omics studies: their use in diagnosis and reclassification of SLE and other systemic autoimmune diseases. Rheumatology (Oxford) 56, i78–i87 (2017).

173. Barturen, G. & Alarcón- Riquelme, M. E. SLE redefined on the basis of molecular pathways. Best Pract. Res. Clin. Rheumatol. 31, 291–305 (2017).

174. Toro- Domínguez, D. et al. Longitudinal stratification of gene expression reveals three SLE groups of disease

activity progression. Arthritis Rheumatol. https://doi. org/10.1002/art.40653 (2018).

175. Coit, P. et al. Genome- wide DNA methylation study suggests epigenetic accessibility and transcriptional poising of interferon- regulated genes in naive CD4+T cells from lupus patients. J. Autoimmun. 43, 78–84 (2013).

176. Jeffries, M. A. et al. Genome- wide DNA methylation patterns in CD4+T cells from patients with systemic lupus erythematosus. Epigenetics 6, 593–601 (2011).

177. Chiche, L. et al. Modular transcriptional repertoire analyses of adults with systemic lupus erythematosus reveal distinct type I and type II interferon signatures. Arthritis Rheumatol. 66, 1583–1595 (2014).

178. Li, Y. & Wu, T. Proteomic approaches for novel systemic lupus erythematosus (SLE) drug discovery. Exp. Opin. Drug. Discov. 13, 765–777 (2018).

179. Bennett, L. et al. Interferon and granulopoiesis signatures in systemic lupus erythematosus blood. J. Exp. Med. 197, 711–723 (2003).

180. Ghodke- Puranik, Y. & Niewold, T. B. Immunogenetics of systemic lupus erythematosus: a comprehensive review. J. Autoimmun. 64, 125e136 (2015).

181. Zhan, Y., Guo, Y. & Lu, Q. Aberrant epigenetic regulation in the pathogenesis of systemic lupus erythematosus and its implication in precision medicine. Cytogenet. Genome Res. 149, 141–155 (2016).

182. Coit, P. et al. Epigenome profiling reveals significant DNA demethylation of interferon signature genes in lupus neutrophils. J. Autoimmun. 58, 59–66 (2015).

183. Tang, Y. et al. MicroRNA-146A contributes to abnormal activation of the type I interferon pathway in human lupus by targeting the key signaling proteins. Arthritis Rheum. 60, 1065–1075 (2009).

184. Coit, P. et al. Renal involvement in lupus is characterized by unique DNA methylation changes in naïve CD4+T cells. J. Autoimmun. 61, 29–35 (2015).

185. Yang, Y. et al. The effect of mycophenolic acid on epigenetic modifications in lupus CD4+T cells. Clin. Immunol. 158, 67–76 (2015).

186. Chafin, C. B., Regna, N. L., Hammond, S. E. & Reilly, C. M. Cellular and urinary microRNA alterations in NZB/W mice with hydroxychloroquine or prednisone treatment. Int. Immunopharmacol. 17, 894–906 (2013).

187. Reilly, C. M. et al. The histone deacetylase inhibitor trichostatin A upregulates regulatory T cells and modulates autoimmunity in NZB/W F1 mice. J. Autoimmun. 31, 123–130 (2008).

188. Barturen, G., Beretta, L., Cervera, R., Van Vollenhoven, R. & Alarcón- Riquelme, M. E. Moving towards a molecular taxonomy of autoimmune rheumatic diseases. Nat. Rev. Rheumatol. 14, 180 (2018).

189. Flint, S. M. et al. Leucocyte subset- specific type 1 interferon signatures in SLE and other immune- mediated diseases. RMD Open 2, e000183 (2016).

190. Banchereau, R. et al. Personalized immunomonitoring uncovers molecular networks that stratify lupus patients. Cell 165, 551–565 (2016). This article presents the first longitudinal study to monitor the transcriptomic profile of patients with SLE and uncover genetic signatures linked to disease manifestations and therapeutic responses in SLE.

191. Peterson, K. S. et al. Characterization of heterogeneity in the molecular pathogenesis of lupus nephritis from transcriptional profiles of laser- captured glomeruli. J. Clin. Invest. 113, 1722–1733 (2004).

192. Stratigou, V. et al. Altered expression of signalling lymphocyte activation molecule receptors in T cells from lupus nephritis patients- a potential biomarker of disease activity. Rheumatology (Oxford) 56, 1206–1216 (2017).

193. Enghard, P. et al. Urinary CD4 T cells identify SLE patients with proliferative lupus nephritis and can be used to monitor treatment response. Ann. Rheum. Dis. 73, 277–283 (2014).

194. Jourde- Chiche, N. et al. Modular transcriptional repertoire analyses identify a blood neutrophil signature as a candidate biomarker for lupus nephritis. Rheumatology (Oxford) 56, 477–487 (2017).

195. Wither, J. E. et al. Identification of a neutrophil- related gene expression signature that is enriched in adult systemic lupus erythematosus patients with active nephritis: clinical/pathologic associations and etiologic mechanisms. PLOS ONE 13, e0196117 (2018).

196. Khamashta, M. et al. CD1067 study investigators. Sifalimumab, an anti- interferon-α monoclonal antibody, in moderate to severe systemic lupus erythematosus: a randomised, double- blind, placebo- controlled study. Ann. Rheum. Dis. 75, 1909–1916 (2016).

N AT U R e R e v I e W S | R h E u M ATO LO Gy

R e v i e w s

v o l U m e 1 5 | J A N UA RY 2 0 1 9 | 47

197. Isenberg, D. et al. Efficacy and safety of atacicept for prevention of flares in patients with moderate- to-severe systemic lupus erythematosus(SLE): 52-week data (APRIL- SLE randomised trial). Ann. Rheum. Dis. 74, 2006–2015 (2015).

198. Ginzler, E. M. et al. Atacicept in combination with MMF and corticosteroids in lupus nephritis: results of a prematurely terminated trial. Arthritis Res. Ther. 14, R33 (2012).

199. Gordon, C. et al. Post Hoc analysis of the phase II/III APRIL- SLE study: association between response to atacicept and serum biomarkers including BLyS and APRIL. Arthritis Rheumatol. 69, 122–130 (2017).

200. Magder, L. S. & Petri, M. Incidence of and risk factors for adverse cardiovascular events among patients with systemic lupus erythematosus. Am. J. Epidemiol. 176, 708–719 (2012).

201. van Assen, S. et al. EULAR recommendations for vaccination in adult patients with autoimmune inflammatory rheumatic diseases. Ann. Rheum. Dis. 70, 414–422 (2011).

202. Andreoli, L. et al. EULAR recommendations for women’s health and the management of family planning, assisted reproduction, pregnancy and menopause in patients with systemic lupus erythematosus and/or antiphospholipid syndrome. Ann. Rheum. Dis. 76, 476–485 (2017).

203. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of an- tiretroviral agents in HIV-1-infected adults and adolescents. tdm-torino http://www.aidsinfo.nih.gov/ContentFiles/ AdultandAdolescentGL.pdf (2011).

204. Freifeld, A. G. et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the infectious diseases society of america. Clin. Infect. Dis. 52, e56 (2011).

205. Dreyer, L. et al. High incidence of potentially virus- induced malignancies in systemic lupus eryhematosus. Arthritis Rheum. 63, 3032–3037 (2011).

206. Bernatsky, S. R. et al. Cancer screening in patients with systemic lupus erythematosus. J. Rheumatol. 33, 45–49 (2006).

207. Watts, N. B. et al. American Association of clinical endocrinologists medical guidelines for clinical practice for the diagnosis and treatment of postmenopausal osteoporosis. Endocr. Pract. 16 (Suppl. 3), 1–37 https://www.aace.com/publications/ guidelines (2010).

208. Grossman, J. M. et al. American College of Rheumatology 2010. Recommendations for the prevention and treatment of gluco- corticoid induced osteoporosis. Arthritis Care Res. (Hoboken) 62, 1515–1526 (2010).

209. Iaccarino, L. et al. Overlap connective tissue disease syndromes. Autoimmun. Rev. 12, 363–373 (2013).

210. Tektonidou, M. G. Antiphospholipid syndrome nephropathy: from pathogenesis to treatment. Front. Immunol. 9, 1181 (2018).

211. Sève, P. et al. Lupus- like presentation of parvovirus B19 infection. Semin. Arthritis Rheum. 34, 642–648 (2005).

212. Verdolini, R. et al. Systemic lupus erythematosus induced by Epstein–Barr virus infection. Br. J. Dermatol. 146, 877–881 (2002).

213. Bezanahary, H. et al. Systemic lupus erythematosus and herpes virus infection: three new observations. Rev. Med. Interne. 23, 1018–1021 (2002).

214. Saravana, S., James, D. W., Abourawi, F., Gupta, P. C. & Samyukta, B. HIV infection mimicking SLE. Clin. Rheumatol. 23, 562–563 (2004).

215. Khosravi, A. R. et al. Severe tinea corporis due to Trichophyton verrucosum mimicking discoid lupus erythematosus. J. Mycol. Med. 22, 92–95 (2012).

216. Arlet, J. B., Capron, L. & Pouchot, J. Visceral leishmaniasis mimicking systemic lupus erythematosus. J. Clin. Rheumatol. 16, 203–204 (2010).

217. Hammoudeh, M. Acute lymphocytic leukemia presenting as lupus- like syndrome. Rheumatol. Int. 26, 581–582 (2006).

218. Yilmaz, M. et al. Histiocytic necrotizing lymphadenitis (Kikuchi–Fujimoto’s disease) mimicking systemic lupus  erythematosus: a review of two cases. Lupus 15, 384–387 (2006).

219. Suwannaroj, S., Elkins, S. L. & McMurray, R. W. Systemic lupus erythematosus and Castleman’s disease. J. Rheumatol. 26, 1400–1403 (1999).

220. Diaz, J. C., Vallejo, S. & Cañas, C. Drug- induced lupus in anti- TNF-alpha therapy and its treatment with rituximab. Rheumatol. Int. 32, 3315–3317 (2012).

221. Soldevilla, H. F., Briones, S. F. R. & Navarra, S. V. Systemic lupus erythematosus following HPV immunization or infection? Lupus 21, 158–161 (2012).

Acknowledgements The authors thank T. Manning for help with revising the English language of the manuscript.

Author contributions All the authors researched data for the article, wrote the arti- cle, reviewed and/or edited the manuscript before submission and provided substantial contributions to discussions of its content.

Competing interests The authors declare no competing interests.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

RelaTed lInks PReCisesADs: http://www.precisesads.eu/

w w w.nature.com/nrrheum

R e v i e w s

48 | J A N UA RY 2 0 1 9 | v o l U m e 1 5