Article Text
Abstract
Objective To evaluate the prevalence, duration and effect on damage accrual of the ‘Lupus Low Disease Activity State’ (LLDAS) in a monocentric cohort of patients with systemic lupus erythematosus (SLE).
Methods We studied 293 Caucasian patients with SLE during a 7-year follow-up period. Disease activity was assessed by SLE Disease Activity Index 2000 (SLEDAI-2K) and SELENA-SLEDAI physician global assessment (PGA), and damage by Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI). We considered the following definition of LLDAS: SLEDAI-2K ≤4 without major organ activity, no new disease activity, PGA (0–3)≤1, prednisone ≤7.5 mg/day and well-tolerated immunosuppressant dosages. The effect of LLDAS on SDI was evaluated by multivariate regression analysis. We also evaluated remission defined as clinical SLEDAI-2K=0 and prednisone ≤5 mg/day in patients treated with/without stable immunosuppressants and/or antimalarials.
Results LLDAS lasting 1, 2, 3, 4 or ≥5 consecutive years was achieved by 33 (11.3%), 43 (14.7%), 39 (13.3%), 31 (10.6%) and 109 (37.2%) patients, respectively. Patients who spent at least two consecutive years in LLDAS had significantly less damage accrual compared with patients never in LLDAS (p=0.001), and they were significantly less likely to have an increase in SDI (OR 0.160, 95% CI 0.060 to 0.426, p<0.001). On average, 84% of patients in LLDAS also fulfilled the criteria for remission.
Conclusions LLDAS was associated with a decrease in damage progression in Caucasian patients with SLE. The majority of patients in LLDAS were in remission, which can largely contribute to the protective effect of LLDAS on damage accrual.
- systemic lupus erythematosus
- outcomes research
- disease activity
Statistics from Altmetric.com
Introduction
Despite the improvement in the diagnosis and treatment of the disease, patients with systemic lupus erythematosus (SLE) continue to be at high risk of morbidity and mortality.1
The efficacy of the treat-to-target approach in improving disease outcomes has recently been demonstrated in rheumatoid arthritis (RA) and other rheumatic diseases.2–6
Indeed, remission and low disease activity (LDA) are currently used as clinical targets in the management of patients with RA.7 Unfortunately, the concept of remission in SLE is less clear than in RA, and a widely accepted definition of remission to be used as therapeutic target in SLE does not exist yet.8 In 2016, an international panel of experts and patient representatives (DORIS—Definition Of Remission In SLE) achieved a consensus, providing a framework for testing different definitions of remission against long-term outcomes, including death and irreversible organ damage.9 In line with these suggestions, we have tested a definition of remission in our cohort of Caucasian patients with SLE.10 11 We demonstrated that remission, defined using clinical, serological and therapeutic criteria, was not rare, and remission lasting at least two consecutive years was associated with a decrease in damage progression.11
Recently, Franklyn et al 12 suggested a definition of LDA in SLE (the lupus low disease activity state, LLDAS), and they tested this definition in their single-centre SLE cohort. They showed that patients followed up for a mean period of 3.9 years (range 0.06–6.9 years) who spent more than 50% of their follow-up time in LLDAS accrued significantly less organ damage than other patients.
The aim of our study was to assess the prevalence of LLDAS, to evaluate its protective effect against damage in our Caucasian SLE cohort and to identify the shortest duration of LLDAS associated with less damage progression in SLE.
Patients and methods
Caucasian patients with lupus prospectively followed at our lupus clinic were eligible for inclusion in the study if they met the SLE American College of Rheumatology (ACR) classification criteria,13 were diagnosed with SLE between January 1990 and December 2008, and had at least three visits per year, no more than 5 months apart, in the 7-year period from January 2009 to December 2015 (an incomplete follow-up due to death did not exclude patients from the study). All patients provided an informed consent before the inclusion in the study.
Information collected at baseline included demographics (age, gender, year of first symptom, year of diagnosis), disease manifestations at baseline and over patients’ disease course, current and previous therapies, complement (C3 and C4) serum levels, antinuclear antibody, antiextractable nuclear antigen antibodies, antidouble-stranded DNA (anti-dsDNA) and antiphospholipid antibodies. The cumulative prednisone dose (g) taken by patients before baseline was calculated.
Clinical and laboratory findings (complete blood cell count, urea and electrolytes, liver function tests, complement and anti-dsDNA serum levels, urinary sediment and 24-hour proteinuria) and data regarding therapy were recorded at each visit.
SLE activity was measured by the SLE Disease Activity Index 2000 (SLEDAI-2K) and Safety of Estrogen in Lupus Erythematosus National Assessment (SELENA)-SLEDAI physician global assessment (PGA, scale 0–3) at each visit. Organ damage was assessed by the Systemic Lupus International Collaborating Clinics/ACR Damage Index (SDI), which was calculated at baseline and annually thereafter. SDI increase was defined as the difference between SDI at the end and at the beginning of the follow-up (‘SDI increase’=‘SDI at the end of 2015’ – ‘SDI at the beginning of 2009’). Antiphospholipid antibody syndrome (APS) was defined according to the revised Sapporo criteria.14
LLDAS was defined as recently reported by Franklyn et al 12: (1) SLEDAI-2K ≤4, with no activity in major organ systems (renal, central nervous system, cardiopulmonary, vasculitis, fever), and no haemolytic anaemia or gastrointestinal active involvement; (2) no new lupus disease activity compared with the previous assessment; (3) a PGA ≤1; (4) a current prednisone (or equivalent) dose ≤7.5 mg/day; and (5) well-tolerated standard maintenance doses of immunosuppressive drugs and approved biological agents. Achievement of LLDAS was determined at each visit.
We evaluated the effect on damage of different durations of LLDAS: 1, 2, 3, 4 and ≥5 consecutive years. In patients who experienced a relapsing-remitting disease, only the longest period of LLDAS was considered in the analysis.
Remission was assessed according to the definitions previously published by Zen et al 11: complete remission was defined as SLEDAI-2K=0 in corticosteroid-free and immunosuppressant-free patients (antimalarials allowed); clinical remission off corticosteroids as serological active clinical quiescent disease (SACQD) according to SLEDAI-2K in corticosteroid-free patients (stable immunosuppressive therapy and antimalarials allowed); and clinical remission on corticosteroids as SACQD in patients on prednisone 1–5 mg/day (stable immunosuppressants and antimalarials allowed).
In this study we classified patients as being in remission or not, irrespective of the level of remission achieved, that is, patients fulfilling any of the three levels of remission were categorised as remitted. Accordingly, remission was overall defined as clinical SLEDAI-2K=0 and prednisone ≤5 mg/day in patients who could be on a stable immunosuppressive and/or antimalarial therapy.
Statistical analysis
A retrospective analysis of the prospectively collected data was performed. Comparison of continuous data with parametric and non-parametric distribution was performed using t-test and one-way analysis of variance with Bonferroni’s post-hoc analysis or by the Wilcoxon’s test and the Kruskal-Wallis test, respectively. Comparison of categorical variables was performed using χ2 test (Fisher’s exact test). Factors with a p<0.2 at univariate analysis were entered into the multivariate model. Backward stepwise multivariate logistic regression was performed with damage accrual considered as a dichotomous dependent variable (ie, SDI increased or not increased during the follow-up), with significance set at 5%. Spearman’s correlation was used to assess the relationship between the duration of LLDAS and damage accrual. Analyses were performed by the SPSS V.23.0 software for Windows.
Results
Two-hundred ninety-three patients were considered in the study. One hundred sixty-nine patients were excluded and the reasons for exclusion were previously detailed.11
During the 7-year follow-up, 33 patients (11.3%) achieved 1 consecutive year LLDAS, 43 (14.7%) achieved 2 consecutive year LLDAS, 39 (13.3%) achieved 3 consecutive year LLDAS, 31 (10.6%) achieved 4 consecutive year LLDAS and 109 (37.2%) ≥5 consecutive year LLDAS. Conversely, 38 patients (13.3%) had never been in LLDAS. Demographic characteristics and damage in our cohort, according to the duration of LLDAS achieved, are reported in table 1, and clinical features and treatment in online supplementary table S1 and supplementary table S2.
Supplementary file 1
Supplementary file 2
Damage more frequently occurred in ocular (17%), neuropsychiatric (16.4%), musculoskeletal (14%) and renal (11.3%) domains, followed by cutaneous (9.5%) and cardiovascular (6.9%) domains; damage in other organs and malignancies were more rarely observed.
The mean SDI increase was lower in patients achieving LLDAS for at least two consecutive years compared with patients never in LLDAS (p<0.001), whereas it was similar in patients with 1-year LLDAS and in those who had never been in LLDAS.
The proportion of patients with damage accrual progressively decreased as the duration of LLDAS increased, both in terms of SDI increase ≥1 or ≥2 (high damage accrual) (table 1).
Moreover, the higher the duration of LLDAS, the lower the increase in SDI (figure 1).
Table 2 reports the characteristics of patients with or without damage accrual during the 7-year follow-up.
In a multivariate logistic regression model including LLDAS and baseline characteristics, an LLDAS lasting 2, 3, 4 and 5 or more consecutive years was protective against damage, whereas age, the use of mycophenolate, a higher cumulative prednisone dose and APS were independent predictors of damage accrual (table 3A). Similar findings were obtained by including in the multivariate model lupus therapies and manifestations ever occurred during the follow-up (see online supplementary table S3).
Supplementary file 3
We also performed a second multivariate analysis, including among the potential predictors of damage a remission lasting at least two consecutive years in addition to LLDAS, since we previously demonstrated its protective effect against damage progression.11 In this model, LLDAS was not independently associated to damage, whereas remission was, meaning LLDAS did not have any additional protective effect over remission (table 3B).
In table 4 the characteristics of patients who achieved or not the LLDAS for at least two consecutive years are reported. As baseline predictors are concerned, patients with a higher SLEDAI-2K, PGA >1, joint and skin involvement, and those treated with methotrexate, ciclosporin and a higher prednisone dose were less likely to achieve an LLDAS lasting two or more consecutive years (table 4).
The multivariate logistic regression model including baseline characteristics showed that a higher cumulative prednisone dose, skin involvement and PGA higher than 1 were the three most significant negative predictors of LLDAS attainment (OR 0.302, 95% CI 0.151 to 0.605, p=0.001; OR 0.333, 95% CI 0.148 to 0.748, p=0.008; OR 0.093, 95% CI 0.151 to 0.605, p<0.001, respectively).
As disease manifestations and treatment over the disease course are concerned, patients with joint involvement, serositis, vasculitis and those treated with immunosuppressive drugs were less likely to achieve an LLDAS for 2 consecutive years. Notably, the majority of patients treated with belimumab achieved LLDAS for at least 2 years (25/30 patients, 83.3%).
We also evaluated the proportion of patients attaining the LLDAS who concomitantly fulfilled the criteria for remission.10 11 Among the 255 patients achieving the definition of LLDAS for at least 1 year during the follow-up, 246 patients (96.5%) satisfied the definition of remission for the same length of time.
Hence, in order to quantify the overlap between the definitions of LLDAS and remission, the proportion of patients achieving LLDAS and meanwhile a remission of the same duration was calculated (figure 2).
Overall, 214 patients (83.9%) experienced a remission being as long as their LLDAS, suggesting a high overlap exists between the two conditions.
Interestingly, remitted patients accrued significantly less damage than did other LLDAS patients (0.59±0.78 vs 0.90±0.89, p=0.021).
Only one death was observed during follow-up; thus, the evaluation of the relationship between LLDAS achievement and mortality was not possible.
Discussion
In the last few years, the beneficial effect of remission on organ damage accrual in patients with SLE has been proven.10 11 15 We demonstrated that prolonged remission, defined as 5 or more consecutive years of no clinical disease activity (SLEDAI-2K=0) in patients treated with antimalarials and/or stable dose of immunosuppressive drugs and/or prednisone ≤5 mg/day, was associated with a decrease in damage progression.10 This finding was confirmed by Tsang-A-Sjoe et al in a multiethnic cohort in the Netherlands.15 In a further study,11 we found that a remission lasting at least 2 years had still a protective effect against new damage.
No shared definition of remission is currently accepted for SLE; thus, the frequency of remission in SLE varies among different cohorts, ranging from 1.2% to 45%.10 11 15–19 This discrepancy led to the search for other measures of outcome, as in other autoimmune diseases, including the so-called LLDAS.
The effect of LLDAS on damage was first evaluated by Franklyn et al 12 who found that among patients followed for an average time of 3.9 years, those who spent more than 50% of their follow-up in LLDAS accrued significantly less organ damage than other patients. This finding was confirmed in the study by Tsang-A-Sjoe et al.15
However, the protective effect on damage of different durations of LLDAS, the shortest duration of LLDAS resulting in decreased damage progression as well as the overlap between LLDAS and remission have not been evaluated yet.
We demonstrated that 2 consecutive years was the shortest duration of LLDAS associated with a decrease in damage progression. Interestingly, since the mean follow-up of patients in Franklyn et al’s study was 3.9 years,12 the 50% of their follow-up time corresponds approximately to 2 years.
Patients who had a higher SLEDAI-2K or PGA, skin and joint involvements, and those treated with a higher cumulative prednisone dose and a higher prednisone dose at baseline were less likely to achieve an LLDAS lasting 2 consecutive years. Similarly, vasculitis, immunosuppressants and high prednisone dose were associated with lack of LLDAS maintenance for at least 50% of follow-up in Franklyn et al’s study.12 In a multinational cross-sectional study, disease duration ≤1 year, history of discoid rash or renal disease, and current elevated anti-dsDNA or hypocomplementaemia emerged as negative predictors of LLDAS achievement at a single point in time.20
Our results are in keeping with the daily clinical practice, since high cumulative doses of prednisone are usually employed in patients with more active and/or refractory disease, who might less likely achieve LLDAS or remission. Skin and joint involvements are usually relapsing-remitting manifestations, which are characterised by a rapid fluctuation of activity over time, and flares can occur shortly after remission is achieved. In addition, although those manifestations are usually considered as mild, it is likely that patients with such organ involvement are treated with less aggressive treatment regimens compared with patients with severe manifestations and are those who more rapidly decrease the prednisone dosage or discontinue induction therapy, being thus at higher risk of early disease flares.
In our study, we considered different periods of time (1, 2, 3, 4, 5 or more consecutive years) instead of the proportion of follow-up spent in LLDAS because such an analysis has more practical implications and can be used in clinical practice to identify patients at higher/lower risk of developing new organ damage.
We evaluated the consecutive time spent in LLDAS and not the sum of separate periods of LLDAS since we previously demonstrated that a shorter consecutive period of remission was as good as a longer cumulative period (3 vs 2 years) in hindering damage accrual over time.11 In fact, it is likely that patients with a relapsing-remitting disease pattern, in whom LLDAS is achieved for short periods and is interspersed by episodes of active disease, require a longer cumulative time of LLDAS to be protected against damage progression.
In addition, we assessed the proportion of patients in LLDAS fulfilling the criteria of remission. In fact, the definition of LLDAS used in the study by Franklyn et al 12 Tsang-A-Sjoe et al 15 as well as in our study included patients who were in LDA but also patients in true remission. This is different from what was defined for RA and other rheumatic diseases,21 22 where the definitions of LDA and remission do not overlap (eg, disease activity score (DAS)-28 in RA identifies remitted patients when DAS-28 is lower than 2.6, while patients in LDA have a DAS-28 between 2.6 and 3.2).
In our cohort we found that the great majority of patients in LLDAS were, actually, in remission. This result is relevant since it is likely that the protective effect of remission on new organ damage significantly contributed to the lower damage accrual observed in the LLDAS group. The finding that remitted patients accrued less damage than other LLDAS patients supports this hypothesis. Moreover, in the multivariate analysis we found that LLDAS did not show any additional protective effect against damage progression over remission.
Unfortunately, the number of patients included in our cohort does not allow an explorative analysis on the effect of LLDAS after exclusion of patients in remission.
Recently, Polachek et al 23 performed a study, in a monocentric cohort of 620 patients, using a definition of LDA different from the Franklyn et al’s definition, namely a clinical SLEDAI-2K <3 (excluding serology), in corticosteroids-free and immunosuppressant-free patients, and a different definition of remission compared with ours, that is, clinical SLEDAI-2K=0 (excluding serology) in corticosteroids-free and immunosuppressant-free patients. They found that the accrual of organ damage in LDA patients was similar to that of patients in remission. However, the definitions of LDA and remission used by Polachek et al are very close to each other, as they both exclude the use of corticosteroids, which are indeed a major determinant of organ damage.
Conversely, the Franklyn et al’s definition of LLDAS and our definition of remission greatly differ in clinical and therapeutic criteria; thus, whether damage accrual is similar in LLDAS and remitted patients is less predictable.
The great overlap between LLDAS and remission observed in our study is not surprising. Indeed, SLE is predominantly a relapsing-remitting disease, which means that its clinical course is characterised by periods of clinical remission intermixed with periods of disease activity. Thus, the disease can be either active or in remission. Accordingly, the LLDAS should be more conveniently applied to patients with a chronic active disease, which is a pattern of SLE activity accounting for a minority of patients with lupus.24
In the multivariate analysis, the protective effect of LLDAS lasting 2, 3, 4 and 5 years on damage accrual was similar, as shown by the overlap in the 95% CIs. This might be due to the relatively short duration of the follow-up in our study. Studies with a longer follow-up could elucidate whether the attainment of prolonged periods of LLDAS has a higher protective effect on damage progression in the long term.
Our study has a number of strengths: we studied a large cohort of patients prospectively and regularly followed for a long period by the same team; we evaluated a range of durations of LLDAS; and we used a validated definition of LLDAS. Moreover, our cohort includes patients treated with biological disease-modifying antirheumatic drugs.
However, our study has also some limitations: we studied only Caucasian patients and we retrospectively analysed data prospectively collected in a single centre. Moreover, this is not an inception cohort and we did not specifically analyse patients with an early disease. Indeed, the purpose of our study was to identify potential predictors of good outcome in a cohort of patients representative of what can be observed in a ‘real life’ lupus clinic.
In conclusion, LLDAS was frequently observed in our SLE cohort and 2 years was the shortest LLDAS duration predictive of better outcome in terms of damage progression. Since a very high proportion of patients in the LLDAS group was also in remission, the remission state could have greatly contributed to the protective effect of LLDAS on damage accrual. A more stringent definition of LLDAS, which does not include the remission state, should be the objective of future studies.
References
Footnotes
Handling editor Tore K Kvien
Contributors MZ contributed to the conception and design of the work, the follow-up of patients, acquisition, analysis and interpretation of data, and she drafted the work; LI followed up patients and gave his contribution in revising the work; MG followed up patients and helped in drafting the work; FS contributed to the acquisition of data; ML followed up patients; AG helped in the analysis of data; LP critically revised the final work; AD led the team who followed up patients, designed the work, interpreted the data, and drafted and revised the manuscript for important intellectual content. All the authors approved the final version of the manuscript and gave their agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Competing interests None declared.
Ethics approval The Ethics Committee of the ’Azienda Ospedaliera-Università degli Studi di Padova', Padova, Italy.
Provenance and peer review Not commissioned; externally peer reviewed.