Introduction
Disease modification is a well-known concept in rheumatology and other therapeutic areas.1–7 However, no widely accepted definition of disease modification exists for SLE. In this article, we explore the core principles of the definitions of disease modification in other conditions, and we propose a conceptual framework for the definition of disease modification in SLE. The demonstration of a medication’s capacity to modify the course of a disease would be a powerful addition to current clinical trial objectives and would ultimately help to improve patient care and clinical outcomes.
Natural history of SLE
SLE is a chronic inflammatory autoimmune disease with fluctuating periods of relative inactivity and disease flares in most patients.8 9 A loss of tolerance to ubiquitous self-antigens in SLE leads to an increase in the number of pathogenic autoantibodies, which drives inflammation and tissue injury, and the onset of clinical manifestations, which may culminate in irreversible tissue damage.10 11 SLE is heterogeneous in its clinical presentations, with multiple organs being affected by varying degrees of severity,9 12 13 which can lead to diagnostic delays and difficulties in evaluating therapeutic outcomes.10 13
Approximately 30%–50% of patients with SLE will develop organ damage within 5 years of diagnosis,14 15 and 50% or more by 10 years postdiagnosis.15 Patients with SLE-associated organ damage have worse health-related quality of life (HRQoL) and increased healthcare costs, morbidity and mortality versus patients with SLE who have lesser or no organ damage.16–23 Damage can develop in several different organ systems, including but not limited to the ocular, musculoskeletal, neuropsychiatric and renal systems.24 Patients with SLE frequently develop lupus nephritis (LN), which can progress to end-stage kidney disease (ESKD), with a 15-year risk ranging from 10% to 30% in patients with severe LN (class III, IV and V) to 44% (95% CI 32% to 56%) in patients with class IV LN.25 26 Organ damage results from potentially modifiable risks such as uncontrolled disease activity/severe flares, exposure to glucocorticoids and/or immunosuppressive therapies and hypertension.27–34 Other known risk factors for accelerated organ damage accrual include pre-existing organ damage, African-American race, older age at SLE diagnosis and male sex.34 35
Treatment goals and outcomes assessments in SLE
According to the European Alliance of Associations for Rheumatology (EULAR) recommendations for SLE management, treatment goals for SLE include control of disease activity; prevention of flares and organ damage; optimisation of HRQoL; and ultimately, prolongation of life. Control of disease activity in SLE should aim to achieve remission or low disease activity and prevent flares in all organs, while minimising treatment-associated toxicities, including exposure to glucocorticoids.36 In patients with LN, the goals of treatment also include preservation of kidney function and prevention of ESKD.37
The complex immunopathogenesis of SLE and heterogeneity of clinical manifestations present challenges in the assessments of outcomes and the designs of clinical trials.10 According to the Outcome Measures in Rheumatology Clinical Trials (OMERACT) initiative, the four core areas that should be assessed in clinical trials of SLE are disease activity, organ damage, HRQoL and adverse events/medication side effects (evaluation of economic cost may also be beneficial, where applicable).38 A summary of key outcome measures and instruments used in clinical trials of SLE and LN is shown in table 1. We will consider the EULAR recommendations for SLE management, the OMERACT areas and established SLE assessments and instruments with the aim of selecting those most suitable for framing a definition of disease modification in SLE that can help inform prescribers and patients. Where appropriate, measures specific to LN also will be considered.
How has disease modification been defined in other disease areas?
The concept of disease modification was first established in rheumatoid arthritis with the introduction of the term disease-modifying antirheumatic drugs in the 1980s and 1990s.39–41 The various definitions of disease modification across multiple therapeutic areas identified in our literature search are summarised in online supplemental table 1.
Across all the disease areas reviewed, affecting a progressive component or the natural course of the disease is central to the definition of disease modification (online supplemental table 1). Improving the signs and symptoms of the disease is a common component within most definitions, which sits alongside modifying the natural progression of the disease but never as a standalone component.1 7 42–44 Two definitions of disease modification therapies referred to those that induce remission or prevent relapses.3 45 The concept of an enduring effect was seen in the definition of disease modification in neurodegenerative disorders and respiratory diseases but not in the other disease areas.46–50 Impact on the disease pathophysiology and quality of life, and evaluation of biomarkers were not consistently included as specific components of a disease modification definition across the disease areas reviewed.7 44 47 48 51 52 Last, in the majority of disease areas, expert-led publications have led the way in setting out possible definitions of disease modification rather than guidance by regulatory authorities.
What are the considerations when defining disease modification in SLE?
Drawing on the definitions from other therapeutic areas, we developed a framework for the definition of disease modification. Two key considerations were incorporated into our formulation, disease manifestations (encompassing signs and symptoms (ie, disease activity) and patient-reported outcomes (PROs)) and disease outcomes (encompassing long-term remission and slowing or stopping organ damage accumulation), and we conceptualise how both considerations may be interpreted to define a long-term meaningful change in the disease course (figure 1).
SLE disease activity is routinely assessed in clinical trials using established instruments, such as the SLE Disease Activity Index (SLEDAI)53 or the British Isles Lupus Assessment Group (BILAG) Disease Activity Index,54 or their modifications (table 1).55–58 Although these measurements are used commonly in clinical trials, they are infrequently used in clinical practice.59 60 The frequency and severity of SLE flares are indicators of global disease control and relevant in both clinical trials and in clinical practice but are not uniformly defined yet.59–62 Reduction in steroid use is a key treatment goal in SLE, due to an increased understanding of the risks associated with long-term exposure.36
Recently, treat-to-target objectives in SLE, in the form of the Definitions of Remission in SLE (DORIS) criteria,63 the Lupus Low Disease Activity State (LLDAS) criteria64 and low disease activity criteria,65 have been associated with reduced disease flares and organ damage accrual in patients with SLE.66–69 Inclusion of these measures, along with flare prevention and glucocorticoid reduction, could be considered key targets in the disease manifestations component of the proposed disease modification definition framework.
The Systemic Lupus International Collaborating Clinics (SLICC)/American College of Rheumatology (ACR) Damage Index (SDI) is a validated measure of organ damage in SLE. However, demonstrating significant effects of an intervention on damage accrual rates requires studies that are longer than 1 year in duration.70–75 In the absence of long-term trials, disease activity, a key driver of organ damage,76 could serve as an interim surrogate measure of organ damage progression to determine whether an intervention is on track to achieve disease modification. Therefore, assessment of both the SDI and key drivers of organ damage should form part of a definition of disease modification in SLE with the direct assessment on organ damage via SDI as the key anchor. In addition, it is particularly important that disease activity is only considered in conjunction with the fewest treatment-associated toxicities, because treatments such as glucocorticoids and immunosuppressants, which reduce SLE disease activity and flares, can promote organ damage progression.29 31–33
In LN, long-term assessment of disease outcomes is focused on slowing or preventing progression to ESKD; and these outcomes can be evaluated by kidney biopsy and other clinical measures of damage, including estimated glomerular filtration rate (eGFR) slope and chronic kidney disease (CKD) staging. Histological findings and chronicity indices determined by kidney biopsy are the gold-standard methods to assess the extent of tissue inflammation and damage.77 78 Histopathology can also be predictive of renal outcome, with fibrinoid necrosis, fibrous crescents and interstitial fibrosis/tubular atrophy ≥25% predictive of progression to ESKD.79 A collaborative effort between the US National Kidney Foundation, the Food and Drug Administration and the European Medicines Agency suggested an eGFR decline of >30%–40% and average eGFR slope reduction (of >0.5–1.0 mL/min/1.73 m2 per year over a 2-year to 3-year follow-up period in clinical trials) as accepted surrogate end points for CKD progression.80 81 Additionally, evaluating SDI at a later timepoint in patients with LN may be valuable to confirm that treatment-associated damage has not taken place.80 81
Improvement in HRQoL and PROs is an important treatment goal in SLE.36 37 82 PROs can be measured using generic instruments such as the 36-item Short-Form Health Survey (SF-36), the Patient Reported Outcomes Measurement Information System item-bank (PROMIS) and Functional Assessment of Chronic Illness Therapy (FACIT)-Fatigue scale, and disease-specific instruments, such as the Lupus Quality of Life (LupusQoL), the Lupus Patient-Reported Outcomes (LupusPRO) and the Lupus Impact Tracker (LIT).83 HRQoL in patients with SLE is complex and can be driven by the disease itself, by comorbid conditions associated with the disease or both; but the precise mechanism by which these processes affect HRQoL is not yet fully understood.84 As such, there is a lack of consensus on the best instrument for their quantification. Given these factors, and taking into account that HRQoL and PROs measures were included in only one disease modification definition (EULAR disease modification definition for RA),43 we suggest that HRQoL and PROs should be supportive elements of the disease modification definition in SLE for the time being.
Last, although definitions of disease modification in neurodegenerative disorders involve targeting the pathogenic mechanisms of the disease and biomarkers, this was not the case with the other disease areas. Indeed, in SLE there are a number of established immunological/serological biomarkers of disease, some of which (anti-double-stranded DNA immunoglobulin (Ig) G, anti-nucleosome IgG, anti-La IgG and anti-Sm IgG, interferon-inducible protein 10, monocytic chemotactic protein 1, macrophage inflammatory protein-3β and complement C3) may correlate with clinical measures of disease activity (using SLEDAI).85 However, the pathophysiology of SLE is complex and heterogeneous among patients and not yet fully elucidated;85 86 as such, there is no single, gold standard biomarker for assessing disease activity, rather a panel of biomarkers would likely be required. Additionally, improvement in clinical and immunological measures do not always occur to the same extent or over the same time course. In fact, immunological responses may mirror, precede or lag behind the clinical responses. Therefore, we felt there would be limited value including pathophysiology and/or biomarkers as a mandatory component of a disease modification definition in SLE; however, it could be considered as a supportive element.
Working definition of SLE-specific disease modification
Based on the framework developed from the definitions of disease modification in other areas, and the treatment goals and outcome measures for SLE, including LN, we suggest the following working definition of SLE disease modification:
'Disease modification in SLE requires minimising disease activity with the fewest treatment-associated toxicities and slowing or preventing organ damage progression (or, in the case of LN, progression to ESKD).'
The evaluation criteria for this working definition is shown in figure 2. The most definitive criteria are slowing or preventing organ damage progression in SLE as judged by no worsening in the SDI and, in LN, by showing a reduction or no worsening in key surrogates of progression to ESKD, that is, kidney histopathology or eGFR decline.
Application of SLE-specific disease-modifying criteria
The disease modification definition for SLE could be applied to both clinical trials and clinical practice (with the notable difference that clinical trials include a comparison arm while clinical practice does not) using specific measures at year 1 to determine an impact on minimising disease activity with the fewest treatment-associated toxicities and at year 5 or later to evaluate slowing or preventing organ damage progression, with interim measures suggested for years 2–5 to determine whether an intervention is on track for achieving disease modification at the 5-year mark, as outlined in table 2.
The 1-year and 5-year timeframes were chosen to assess early markers and confirmed disease modification, respectively, based on thresholds for assessment of disease activity and organ damage in clinical trials and longitudinal studies.38 80 Examples of tools to be used to assess disease activity were provided, without specifying a single tool, to allow for evolution of this definition and how it can be assessed. A range of steroid thresholds are suggested to reflect the range recommended in treatment guidelines and criteria for remission and low disease activity state.36 63 64