Comparison of three different definitions of low disease activity in patients with systemic lupus erythematosus and their prognostic utilities

Abstract

Objectives

Three definitions of low-level disease activity in patients with SLE have been proposed by different groups. These include minimal disease activity (MDA), low disease activity (LDA) and the lupus low disease activity state (LLDAS). We investigated the performance of these definitions in SLE patients.

Methods

We recruited 299 SLE patients who were followed up annually for 4 consecutive years. We compared the three definitions of low disease activity via longitudinal analysis; we used a generalized, linear-mixed effects model and generalized estimating equations.

Results

The LLDAS was significantly associated with a lower SLICC/ACR damage index (β coefficient=−0.064, 95% CI: −0.129, −0.002, P=0.050), reduced flare (odds ratio = 0.090, 95% CI: 0.034, 0.239, P<0.001), an improved SF-36 physical component score (β coefficient=0.782, 95% CI: 0.046, 1.519, P=0.037), and an improved SF-36 mental component score (β coefficient=1.522, 95% CI: 0.496, 2.547, P=0.004). Neither the MDA nor the LDA were associated with these variables.

Conclusion

The LLDAS definition performs better than the MDA and LDA definitions, showing that LLDAS is associated with less organ damage and flare, and a better quality of life, during follow-up.

Introduction

SLE is a chronic autoimmune disease characterized by heterogeneous clinical manifestations and a highly fluctuating disease course [1]. The treat-to-target approach has been proposed to prevent cumulative damage and improve the quality of life in patients with systemic autoimmune diseases including SLE. Suggested SLE targets include remission and low disease activity. Previous studies found that remission was extremely rare and usually not sustained [2, 3]. Recent studies, conducted using a multinational cohort and the Hopkins lupus cohort, showed that low disease activity was easier to achieve in SLE patients [4, 5]. Thus, low disease activity may be a more realistic target. All of minimal disease activity (MDA) [6], low disease activity (LDA) [7] and the lupus low disease activity state (LLDAS) [8] have been proposed to define low-level SLE activity. To the best of our knowledge, no head-to-head comparison of the three definitions has been performed. Therefore, the aim of this study is to establish agreed-upon definition of low disease activity by comparison of three definitions and identify its association with prognostic value in Korean patients with SLE.

Methods

Study design and population

We accessed the database of the KORean lupus NETwork (KORNET) registry. KORNET is a multicentre hospital-based registry that seeks to evaluate prospectively the characteristics and clinical outcomes of Korean patients with SLE. From January 2014 to December 2018, we enrolled patients diagnosed with SLE using the 1997 ACR revised criteria [9]. The KORNET registry contained data on 543 patients who were followed up. Of these, 76 were lost to follow up, 37 withdrew their informed consent, and 131 were followed up for <4 years. We reviewed the annual assessments of the remaining 299 patients over four consecutive years. Informed consent was obtained from all participants prior to enrolment, and the study was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki and later amendments thereof. The study was approved by the Institutional Review Board of Chonnam National University Hospital (no. CNUH-2015–250).

Definitions of low disease activity

We compared the three definitions of low disease activity: MDA, LDA and LLDAS. The MDA is based on a clinical SLEDAI-2000 (SLEDAI-2K) score ≤1 (excluding active serology, thus ignoring anti-dsDNA positivity and/or low complement levels) maintained using antimalarials, standard doses of immunosuppressants (not exceeding the maximum recommended doses) and low-dose prednisolone (≤5 mg/day) [6]. The MDA allows only hematological manifestations, such as isolated leukopenia or thrombocytopenia, which must be attributable to lupus and not to secondary causes. The LDA is based on a SLEDAI-2K score ≤2 (with or without positive serology, based on the presence of only one of rash, alopecia, mucosal ulcers, pleurisy, pericarditis, fever, thrombocytopenia or leukopenia) in patients on antimalarials only, but not glucocorticoids or immunosuppressive agents [7]. The Asia-Pacific Lupus Collaboration defines the LLDAS as follows:

• SLEDAI- 2K score ≤4, with no adverse renal, CNS, cardiopulmonary, gastrointestinal or vascular activity;

• no fever, and no haemolytic anaemia;

• no new lupus disease activity compared with the previous assessment;

• Safety of Estrogens in Lupus Erythematosus National Assessment (SELENA) SLEDAI Physician Global Assessment (PGA) (scale 0–3) of ≤1;

• a current prednisolone (or equivalent) dose ≤7.5 mg daily; and

• well-tolerated standard maintenance doses of immunosuppressive drugs and approved biological agents, excluding investigational drugs [8].

We evaluated all participants fulfilling any of the three definitions of low disease activity. We assessed all patients at each annual cohort visit in terms of whether they exhibited low disease activity at those times. Sustained low disease activity was defined as low disease activity over 4 consecutive years.

Clinical outcomes

We assessed the sociodemographic and clinical features, laboratory findings and medications of each patient annually. Clinical manifestations (mucocutaneous, musculoskeletal, renal, neurological, and haematological disorders and serositis) were assessed as suggested by the 1997 revision of the ACR criteria for SLE [9]. Routine laboratory findings and the levels of autoantibodies and anti-phospholipid antibodies were also reviewed. The outcome measures included flare, SLEDAI-2K and PGA scores, the SLICC/ACR damage index (SDI), the short form-36 (SF-36) instrument, and the Beck Depression Inventory (BDI). Disease flares in the time since the previous visit were investigated [10]. Disease activity was measured using the SLEDAI-2K index [11], and PGA of disease activity was assessed on a scale of 0–3. Irreversible organ damage was measured using the SDI [12]. Quality of life was assessed using the Medical Outcomes Study 36-item short-form health survey (SF-36); this generic instrument has been validated in clinical trials [13]. The Korean version of the BDI was used to assess comorbid depression; the instrument contains 21 multiple-choice questions [14].

Statistical analysis

Data processing and statistical analyses were performed using STATA v15 (StatCorp., Austin, TX, USA). Continuous variables are presented as medians with interquartile ranges. Descriptive statistics were used to compare low disease activity frequencies (yes/no) at each follow-up cohort visit based on the three definitions. In other words, we assessed all participants at all visits in terms of whether each of the three definitions of low disease activity were met. We included all possible cases. As we analysed time-dependent variables including disease activity, the SDI and BDI scores, flare, and SF-36 data, we used analyses that allowed for changes over time. We employed generalized estimating equations (GEEs) to identify longitudinal associations between each definition of low disease activity and clinical outcomes over the 4 years. Binary clinical outcome such as flare was subjected to GEE logistic regression analysis and continuous clinical outcomes such as the SDI, BDI and physical and mental component scale scores to GEE linear regression analyses. Regression coefficients (β values) with 95% CIs were calculated for all scores; and odds ratios (ORs) with 95% CIs for flare. A P-value<0.05 was considered to indicate statistical significance.

Results

We assessed 299 SLE patients annually for 4 consecutive years. Their baseline characteristics are listed in Table 1. The median age at enrolment was 40.0 years (range 32.0–49.0 years) and 91.7% were female. The median disease duration was 89.0 months (range 37.0–165.0 months).

The frequencies of low disease activity, both incidental and sustained, yielded by the three definitions throughout follow-up are listed in Table 2. The MDA frequencies from year 1 to year 4 were 29.1%, 33.8%, 20.7% and 22.7%. The LDA frequencies from year 1 to year 4 were 4.0%, 4.0%, 3.3% and 2.7%. The LLDAS frequencies from year 1 to year 4 were 70.2%, 72.6%, 59.5% and 66.6%. The numbers of patients exhibiting sustained low disease activity over four consecutive years were 11 (3.7%) by the MDA, 2 (0.7%) by the LDA and 86 (28.8%) by the LLDAS.

Table 3 shows the longitudinal associations over 4 years between time-varying, low-level disease activity and clinical outcomes, analysed using the GEE method. The LLDAS was significantly associated with a lower SDI score (β=−0.064, 95% CI: −0.129, −0.002, P=0.050), reduced flare (OR=0.090, 95% CI: 0.034, 0.239, P<0.001), an improved SF-36 physical component score (β = 0.782, 95% CI: 0.046, 1.519, P=0.037), and an improved SF-36 mental component score (β = 1.522, 95% CI: 0.496, 2.547, P=0.004). Neither the MDA nor the LDA were associated with the SDI score, flare or SF-36 score. There was no association between any definition and depression.

Discussion

Remission is the target of SLE treatment, but sustained remission is very rare. Thus, low-level SLE activity may be an acceptable and more realistic target. Although three distinct definitions of low-level disease activity have been proposed, no head-to-head comparison has yet been performed. Here, we show that the LLDAS definition of low-level SLE disease activity was better than the MDA and LDA definitions; LLDAS was associated with lower SDI scores, reduced flare, and improvements in both of the SF-36 components during 4 years of follow-up.

There were three distinctive definitions of low disease activity in SLE, and applicability of these definitions was been repeated in several independent cohorts in Europe and North America. Franklyn, et al. of the Asia-Pacific Lupus Collaboration, developed the LLDAS concept based on expert opinion [8]. Attainment of LLDAS status for >50% of the follow-up period was significantly associated with lower risk of accrued damage, and similarly, failure to achieve LLDAS by 6 months was an independent predictor of early damage [4, 15–18]. Golder et al. reported that attainment of LLDAS at any time was associated with later reductions in flare and organ damage [5]. In a Norwegian inception cohort, patients who were of LLDAS status for ≥50% of their follow-up time exhibited less organ damage and mortality [19]. The LDA was developed by Polachek et al. and LDA patients did not differ from those in remission in terms of the adjusted mean SLEDAI‐2K score, organ involvement, SDI score, mortality or required therapies after 2 and 10 years of follow-up [7, 20]. After 2 and 4 years of follow-up, LDA patients had a lower adjusted mean SLEDAI-2K score, less major organ involvement, a lower SDI score, lower mortality and required less therapy compared with those exhibiting high disease activity, suggesting that sustained LDA was an acceptable treat-to-target outcome [7]. The MDA definition was introduced by Zen et al. and as was true of LDA patients, MDA patients were comparable in terms of damage accrual to those who achieved clinical remission after 7 years of follow-up [6]. MDA attainment was associated with less mortality during follow-up [21]. Notably, when the three definitions were compared in a prospective cohort of ethnically homogeneous Korean lupus patients, patients who achieved LLDAS status exhibited less organ damage, reduced flare and an improved quality of life compared with those who achieved MDA or LDA status.

There are several possible explanations as to why the three definitions capture different outcomes in lupus patients. First, differences exist across three definitions in terms of level of acceptable activity as expressed by permissible clinical manifestations, active serology and acceptable treatment. For disease activity measured by the SLEDAI-2K, the MDA and LDA definitions accept scores of 1 and 2, respectively, but the LLDAS definition accepts scores up to 4. In terms of permitted treatment, the LDA definition allows only antimalarials during follow-up, but the LLDAS and MDA definitions permit immunosuppressive agents and low-dose prednisone. In terms of corticosteroid dosage, the MDA definition allows prednisone up to 5 mg/day but the acceptable LLDAS dose is up to 7.5 mg/day. Thus, the LLDAS is the least stringent definition in terms of disease activity and treatment. As the MDA and LDA definitions approximate to clinical remission, the disease activities and treatment requirements of MDA and LDA patients are similar to those of patients in remission [22]. We found that the proportions of patients who sustained MDA and LDA status during the 4-year follow-up period were only 0.7 and 3.7%. In contrast, 28.8% sustained LLDAS status. It is easily expected that the number of patients who satisfied the definitions of MDA and LDA were too small that could fail to get enough statistical power. Thus, achievement of MDA and LDA status did not translate into a lower SDI score, reduced flare or an improved quality of life in the present study. In this regard, LLDAS can be a reasonable therapeutic target and seems to have flexible availability in real clinical practice.

Second, the MDA and LDA definitions are based on the clinical SLEDAI-2K scores rather than the overall SLEDAI-2K scores. The clinical SLEDAI-2K index excludes active serology data, such as low complement levels or abnormal titers of anti-dsDNA antibody. The MDA and LDA definitions focus principally on clinical features. Steiman et al. [23] showed that over half of patients who were serologically active but clinically quiescent developed flares during 3 years of follow-up. Thus, it is possible that patients who satisfied the MDA and LDA definitions experienced flares of varying severity during follow-up. In other words, maintenance of MDA and LDA status may not be associated with lower SDI scores, reduced flare or an improved quality of life. As sustained LLDAS status was associated with favourable clinical outcomes during follow-up in the present study, attainment of LLDAS status is a practical treatment goal minimizing organ damage, optimizing health-related quality of life, and improving long-term survival.

Our study had several strengths. First, this is the first head-to-head comparison of the three definitions in an established longitudinal cohort who were ethnically homogeneous. Second, we analysed organ damage, flare, and quality of life using time-dependent parameters that considered changes over the follow-up period. We annually assessed all participants in terms of each definition of low disease activity for 4 years. We obtained SDI and SF-36 scores and assessed flare when exploring the effects of MDA, LDA and LLDAS status.

In conclusion, we explored the impacts of three definitions of low disease activity on long-term SLE outcomes including damage accrual, flare and health-related quality of life. The LLDAS definition performed better than the MDA and LDA definitions in terms of reducing organ damage and flare and improving the quality of life. The LLDAS definition will become more widely used than the MDA and LDA definitions in both routine clinical practice and randomized trials evaluating newer drugs.

Acknowledgements

We thank the patients and their families for their participation.

Funding: This study was supported by a grant (BCRI20015) from Chonnam National University Hospital Biomedical Research Institute.

Disclosure statement: The authors have declared no conflicts of interest.

References