4.  Patient selection/diagnostic criteria

4.1. Systemic Lupus

Participating patients should have a definite diagnosis of SLE based on generally accepted classification criteria, i.e. the revised American College of Rheumatology classification criteria or the SLICC SLE classification criteria.  It is advisable to adhere to one classification set throughout the clinical development programme. Considering that SLE can have a wide range of manifestations and affected   patient populations can be diverse, it is encouraged that a broad  spectrum of patients compatible with the objectives of the planned clinical trial should be enrolled unless a specific subset or subsets of SLE  patients is planned to be targeted (e.g. renal lupus).

In all cases, demographics and baseline disease  characteristics including duration of the disease, disease manifestations, previous and concomitant therapies (including those not directly aimed at SLE, but which could for example alter the extent of organ damage, e.g drugs associated with photosensitivity, thrombocytopenia etc.), should be carefully recorded at the beginning of the study.

The serologic markers such as positivity for ANA, anti-dsDNA and complement levels should additionally be considered at study entry.

Patients should also be characterised with respect to disease activity and the severity of the disease.

In order to demonstrate a significant reduction in disease activity it is recommended that patients have a clinically important and sufficient level of disease activity despite prior treatment, at least  moderate  to severe active disease defined by e.g. baseline SLEDAI >6.  Care should be taken to distinguish disease activity at enrolment from the level of damage and functional disability reached by the patient due to the course of the disease prior to baseline. Therefore, it should be clear how the activity and severity of disease has been measured and the primary data collected should be carefully recorded.

In addition to the level of disease activity at enrolment, documentation of the frequency of flares, e.g. the number of well documented flares for a period of 6-12 months prior to enrolment, should be recorded.

It should be ensured that different study arms are balanced for patients’ characteristics and prognostic factors, including baseline disease activity, ethnicity and background therapy (e.g. glucocorticoid use    above a given dose). If any baseline factors may in themselves significantly influence prognosis, prior  stratification according to these factors should be considered.

Additionally, if certain medication is not allowed, a drug free interval should be specified. In the case that a prerequisite for enrolment is a discontinuation of certain medications, the reason for discontinuation e.g. lack of efficacy, intolerance or adverse reactions, and the necessary wash out intervals should be clearly defined and justified in the protocol.

5.  Efficacy assessment

5.1. Primary outcomes in SLE

The selection of the primary endpoint will depend on the objective of the clinical study and may be   generally aimed at the control of disease activity and/or prevention of long-term damage. Given the heterogeneity of SLE manifestations, supportive evidence derived from secondary endpoints is of particular interest in this clinical setting to fully characterise the treatment effect.

5.1.1.  Control of the disease activity

5.1.1.1. Reduction of the level of disease activity

The therapeutic goal of any given treatment intended for chronic use is to induce a clinically relevant reduction in the activity of the disease that is maintained in the long term.

Several validated scales of disease activity are available. Thus far, SLEDAI and BILAG have been shown to be sensitive in clinical trials. However, in view of the complexity of SLE, measurement of disease activity by a single index alone is considered insufficient to fully describe the therapeutic effect  in individual patients. It should be demonstrated that an effect on reduction in global disease activity is not offset by worsening in any specific organ system. Therefore, it is recommended to assess the effect on disease activity preferably through validated composite indexes, (i.e., SLE Responder Index [SRI] or BILAG-based Composite Lupus Assessment [BICLA]) , which both include a measure of global disease activity (by SELENA-SLEDAI), specific organ system involvement (BILAG) and overall subject’s condition (Physician’s Global Assessment).

The preferred main analysis should rely on the differences in the rate of responders.  Responder criteria should be pre-defined, e.g. the target outcome could be remission, or a major clinical response, as defined by SLEDAI score of ≤2 or SRI ≤4, BILAG score of C, PGA of improvement, with or without specific laboratory tests where relevant. Complete clinical remission is defined by complete absence of disease activity measured by disease activity indices in patients who do not require any ongoing lupus specific therapy. For patients on steroids, including steroid dependent patients, the goal of treatment is steroid-free or at least to achieve a low-steroid dose to maintain remission.

If partial response is targeted, defined as a clinically relevant response not fulfilling criteria for major response, this should be accompanied by a favourable safety profile, and/or a significant reduction of steroid use (see section 5.2.1). The clinical relevance of response criteria should be adequately justified, chosen before the study is initiated and thresholds should be thoroughly predefined.

The results should also be presented by both the absolute and the percentage change of the selected  index/ composite between baseline and the end of the trial when possible, depending on nature of the index/composite.

There are several other scales available or underdevelopment to assess disease activity in SLE. Other scales should not be used as a primary endpoint, unless they are fully validated and generally accepted. It is recommended to seek scientific advice prior to the studies if there is deviation from the SRI as primary endpoint.

Patients should be followed up and assessed regularly in order to evaluate the response trends and establish the start of the effect, the peak and maintenance of effect.

The time point chosen for the evaluation of the effect on disease activity will depend on the time it takes the study drug to achieve an optimal effect, on the severity of the disease and whether it is  intended for acute or chronic use. In some cases a response, informed by pharmacodynamic properties, may be considered achievable within 4 to 8 weeks from the start of treatment.  However, for drugs intended for chronic use, control of disease activity should be determined over the longer    term, i.e. at least 1 year.

5.1.2.  Prevention of flares

The aim of any study drug intended for chronic use is to maintain the clinical response previously achieved by demonstrating an effect on the prevention of flares (decrease frequency/severity).    Therefore, this should be measured either as a primary endpoint or as a key secondary endpoint.

Criteria for flares should be predetermined in the protocol. The characteristics of the flare include a clinically significant measurable increase in disease activity in one or more organ systems. It is most    commonly a temporary event and usually there would be at least consideration of initiation or increase in treatment.

The flare is reflected by a clinically meaningful increase in an accepted disease activity score, for example an increase in SLEDAI-2K score ≥ 4 points, an increase in SELENA-SLEDAI score of ≥ 3 points or 1 new category A or 1 or 2 new category B items on the BILAG score. Other tools and/or definitions  can be used if generally accepted and well justified.

Either, the time to a new flare or the frequency/annual rate of flares according to accepted criteria should be measured. If the time to a new flare has been chosen as a primary endpoint, the rate of flares over appropriate time points should be included as a secondary endpoint. An evaluation of the frequency of flares should normally be made over a period of at least one year. The protocol should establish the requirements to consider changes in disease activity as a new flare and not part of the previous episode. In addition, maintenance of response, i.e. defined by differences in the proportion of patients in different study arms who remain flare free over at least 12 months is considered as a relevant primary or secondary endpoint. A study duration of at least 1 year is usually needed to demonstrate the effect on flares.

5.1.3.  Prevention of long term damage

Accumulated multi-system chronic organ damage as measured by the SLICC/ACR damage index is suitable to use in studies enrolling patients particularly those with short duration of disease and without pre-existing damage,  as it is more difficult  to evaluate differences in damage accrual if the  population enrolled has highly variable baseline damage. As the evaluation of damage accrual will be clearer in those with low baseline damage, it is recommended to stratify by baseline damage.

Manifestations should be recorded as damage only if they develop at or after the diagnosis of lupus, provided they fulfil the list of definitions, and irrespective of whether they are considered to be directly

attributable to the disease. Damage accrual is usually recorded if present over 6 months or associated with immediate pathological change indicative of damage. Therefore the clinical trial should be long enough (at least 12 months) for damage to occur and persist for 6 months. Using a SLICC/ACR damage index may be problematic when a new study drug is associated with toxicities not listed in the Damage Index used to. In this event, other instruments to assess damage could potentially be used,    however this should be discussed with relevant regulatory authorities prior to commencing trials. For a therapy that has efficacy in reducing disease activity, long-term follow-up of patients in a controlled extension would be required to demonstrate efficacy for reduction of damage.

5.2. Other relevant secondary endpoints for SLE

When a composite endpoint that consists of multiple indices (e.g SLE Responder Index) is used as a  primary outcome measure to assess the efficacy of the drug, components of this composite endpoint should be analysed separately as secondary outcomes and described alongside the result for the composite outcome. Relevant endpoints not selected as primary should be included as key secondary endpoints.

5.2.1.  Decrease in steroid dose

The concept of steroid-sparing is a key variable to consider in trials assessing add-on and maintenance effect during which the aim is to reduce the dose or even discontinuing steroids without precipitating a flare. The timing of altered doses in relation to the final outcomes needs to be considered. In the protocol it should be clear at which time point dose changes within predefined limits would be allowed. The protocol should also specify if administration of other forms of steroid including parenteral, intra-   muscular or intra-articular steroids are allowed.

The efficacy evaluation for steroid tapering should be based on the percentage of patients whose average prednisone (equivalent) dose can either be discontinued while maintaining control of disease activity or at least reduced by a clinically relevant magnitude according to different stringent pre-specified criteria , e.g. subjects whose prednisone equivalent dose was 15 mg/day at baseline and reduced to ≤7.5 mg/day., without any flares for at least the final 3 months in a trial lasting one year. Reductions should have meaningful clinical implications

5.2.2.  Patients and investigators reported outcomes

Quality of life

Health related quality of life (HRQoL) is known to be compromised in lupus patients and appears to be an independent outcome measure that does not necessarily correlate strongly with disease activity or  organ damage.  Given the fact that HRQol is an independent outcome in Lupus patients, it is therefore important to consider assessing the impact of the condition on the patient’s HRQoL to have a complementary source of information beyond what is captured by measures of disease activity and organ damage.

Medical Outcome Study Short Form 36 (SF-36) has widely been used to assess physical, psychological and social impact of chronic disease like lupus. An additional tool available to assess the impact of disease on the patients includes the Work Productivity and Activity Impairment Lupus score.

Lupus specific instruments include the Lupus Quality of Life (Lupus QoL), SLE symptom checklist and SLE Quality of Life (SLE QoL). As these instruments have not been validated in clinical trial settings and their correlation with SF-36 is variable, it is prudent to use these instruments together with SF-36.

Fatigue is a major concern for adults with SLE affecting more than 90 percent of the lupus patient population. Many instruments have been developed to measure fatigue severity and its impact in SLE,   including the fatigue severity scale (FSS) which is most commonly used and correlates moderately with the 8 scales of SF-36, the FACIT fatigue or the Brief Fatigue Inventory (BFI). Other alternatives might   be used provided they are validated and generally accepted. Changes in physical function by e.g. ADL may also be of interest.

5.2.3.  Biomarkers

Although a large number of novel biomarkers have been studied in lupus, none of them have been  rigorously validated in longitudinal studies and in different ethnic cohorts. Furthermore, a candidate biomarker or combinations of them will unlikely substitute for conventional clinical parameters for    monitoring the disease course. However, such biomarkers when used in combination with clinical parameters may improve efficiency of confirmatory trials with respect to patient selection, dose optimisation, and identification of drop outs with the future aim of developing more targeted treatments. It is therefore advised that identification and inclusion of biomarkers is incorporated as an integral part of the drug development programme.

6. Strategy and Design of Clinical Studies

6.1. Exploratory studies

6.1.1.  Pharmacokinetics

The pharmacokinetic properties of the medicinal product should be thoroughly investigated in accordance with relevant guidelines regarding interactions, special populations (elderly and paediatric,  renal and hepatic patients), and specific quality aspects (locally applied drugs, proteins and monoclonal antibodies).

6.1.2.  Dose response studies

For dose response studies, ICH E4 guidance Dose-Response Information to Support Drug Registration should be considered. Evaluation of multiple doses is recommended. Should specific indications targeted at specific patients be sought, efforts should be made to find different doses and treatment intervals according to the respective patient characteristics (i.e. severity, organ involvement). For products being investigated in 2 or more autoimmune diseases extrapolation of dose-finding across indications could be acceptable subject to adequate justification.

Placebo controlled, randomised, double blind and parallel group design is recommended. Duration of the dose finding study depends on the lupus patient profile (e.g. severity of organ manifestations),   chosen endpoints and mode of action of the medication.

6.1.3.  Interactions

Interaction studies should be performed in accordance with the existing guidelines. Efficacy and safety implications of concomitant drugs likely to be co-administered in clinical practice (e.g. glucocorticoids, immunosuppressant’s, NSAIDs) should be evaluated.

6.2.  Therapeutic Confirmatory Studies

6.2.1.  Systemic Lupus

Study design, outcome measures and duration should be appropriately chosen and justified with regard to the mode of action, magnitude and time course of effect of the test drug.

Although SLE has a fluctuating course, induction and maintenance therapy are not considered as separate treatment modalities of SLE treatment in clinical practice -this in contrast to lupus nephritis, where treatment of acute flares is separated from  chronic maintenance treatment.

Double blind, parallel group, randomised trial design is recommended.  Superiority trial design against an active comparator or placebo is preferred. Non-inferiority studies could only be accepted provided   that the selected comparator could be justified on the basis of a well-established efficacy, and an appropriately justified non-inferiority margin could be predefined.  Such comparative studies must have assay sensitivity. Alternative designs might be considered but it is recommended to discuss the design and planned data analysis methods with regulatory authorities before initiating the studies. Placebo controlled trials are acceptable provided that placebo is given in add-on to standard of care therapy, unless otherwise justified. In placebo controlled add-on design setting the background treatment becomes of particular relevance. In order to avoid sub-optimal treatment in the control group of SLE patients, predefined adjustments in the background treatment should be planned, allowed and presented in the protocol. Escape provisions to an alternative standard-of-care regimen for patients who worsen during the study can be included and should be predefined.

The possibility of including an active comparator in the study design could be considered and predefined in the protocol.  The three-arm placebo- and active-controlled study would  aim to demonstrate that the test product is superior to placebo and allow putting results into perspective; It will address the real contribution of the new substance and could give clues for its suitability as first   line treatment in some patients. Possible primary endpoints include clinically relevant sustained reduction in disease activity. Alternatively, efficacy in the control of disease activity can be demonstrated by a significant effect in    the prevention of moderate-severe flares.  Considering the fluctuating nature of SLE, the duration of a trial aimed for the control of disease activity should be at least 12 months. Other clinical trial designs that provide enhanced information to understand long-term efficacy and   persistence of effect might be considered, e.g. a randomised withdrawal against placebo once short- term efficacy has been shown.

6.2.3.  Background medication

Changes in background medications that are used to treat patients with SLE can obscure detection of a treatment effect with the study drug. Therefore, background therapy should be standardised and stable as far as possible without compromising optimisation. Patients’ needs during the trial should be addressed appropriately. Certain common practice modifications of background therapy could be allowed; these modifications should be well defined and carefully documented in the protocol (this includes also non-SLE medication, e.g. ACE inhibitors). If medication changes are substantial based on uncontrolled disease, then patients can be rescued and considered for the purpose of the primary efficacy endpoint as treatment failures. The trial should include predefined escape conditions to allow switching to “rescue medication” when the patient fails to improve or the condition worsens. The choice and terms of rescue medication should be predefined in the protocol.

7. Juvenile-onset SLE

Although direct comparison with adult-onset disease is sparse due to the low incidence of juvenile-onset SLE (less than 1/100,000), there is evidence to suggest that juvenile-onset SLE patients (disease onset before 18 years) display some differences in their disease profile. Compared with adult- onset SLE populations there is an increased male-to-female ratio, a higher prevalence of nephritis and  CNS involvement and faster accrual of damage in juvenile-onset SLE. This would often necessitate aggressive treatment and sustained need for steroids. Therefore specific instruments to assess disease outcome are needed that would on the one hand take into consideration disease course and aggressive therapy and on the other hand take into consideration the growing and developing paediatric patient whose perception of disease can be very different from adults and depend on the age group. Patient Reported Outcome (PRO) measures may require validated adaptations for use in paediatric subjects, such as incorporating symbolic visual anchors and/or age-appropriate text in symptom rating scales.

The Paediatric Rheumatology International Trials Organization (PRINTO), has developed a core set of five domains for the evaluation of overall response to therapy in juvenile-onset SLE. These domains   include the following:

•     Physician’s global assessment of disease activity;

•     A global disease activity measure (e.g. European Consensus Lupus Activity Measure (ECLAM),

Systemic Lupus Erythematosus Disease Activity Index (SLEDAI), Systemic Lupus Erythematosus Activity Measure (SLAM), British Isles Lupus Assessment Group (BILAG), or other global disease activity measures deemed appropriate for clinical trials)

•     24-hour proteinuria. Alternatively the spot urine protein: creatinine ratio on first morning void urine sample is considered a valid measurement.

•     Patient’s/Parent’s global assessment of the overall patient’s wellbeing

•     Health-related quality of life assessment (Child Health Questionnaire physical summary score)

According to the PRINTO/ACR criteria patients are classified as responders if they demonstrate at least 50% improvement from baseline in any 2 among 5 core set measures with no more than 1 of the remaining worsening by more than 30%. The PRINTO/ACR criteria can be applied to all subtypes of    juvenile SLE including trials specially designed for patients with renal involvement. Other composite    indices that account for both improvement and worsening of disease manifestations in different organ systems, such as the SRI or BICLA, as well as time to SLE flare may also be acceptable primary efficacy endpoints. Paediatric adjusted parameters (e.g GFR, blood pressure adjusted to the age, sex and height of the patient) should be used when evaluating clinical activity of the disease In long term follow up the accrual of damage caused by the disease could be evaluated using SLICC/ACR damage index. Pharmacokinetic, dose finding and where relevant vaccination/immune response studies should be performed in children from 5 to less than 12 years old. Where possible dosing for adolescents may be extrapolated from adults. There is no need for development of medicines for SLE in children under 5   years of age as the disease is extremely rare in that age group. Safety may not be fully extrapolated from adults: however it is not realistic to accumulate sufficient   information on safety in pre authorisation studies in children. Long term post authorisation studies or establishment of patient registries are necessary.

8.  Elderly

While onset of SLE is generally between the ages of 15-45 years, the improved survival of patients with SLE over the last 20 years  and in addition cases of late onset SLE means that older patients should be included in clinical trials of adult SLE. Study data should be reported separately for patients aged 65-74, 75 and older where available.

9. Clinical Safety Evaluation

Specific adverse events to be monitored

The analyses of safety data should particularly focus on specific adverse effects related to the mode of action or risks known for the specific substance class. These specific adverse effects might occur after  drug discontinuation and should be evaluated and documented for an appropriate period post study. As the risk of malignancy, infection and cardiovascular events is greater in SLE patients, this should be specifically monitored. As the kidney is an important target organ for SLE, potential impact of the new  agent on renal function or, in turn, any influence of renal impairment on drug elimination should be adequately monitored. Adverse events related to common organs/systems involved in SLE should also be closely monitored. Long term follow-up data must be available. The extrapolation of data from the general safety database for organ specific conditions should be thoroughly justified.

Long-term safety

SLE is a chronic disease and most systemic drugs will need to be approved for long-term treatment or chronic repeated use. Thus safety assessment should be consistent with standard CHMP requirements for safety data applicable to long-term treatments. Importantly, a risk management plan should be submitted that includes measures for provision of long term data post authorisation to assess the   development of related malignancies. For further identification of rare adverse events associated with new therapies, intensive safety evaluation during randomised trials might contribute but long-term follow-up in a large population will be needed.